Model parameters: d_model 768 ffw_size 3072 kv_size 64 n_heads 12 n_layers 15 Megatron-DeepSpeed/pretrain_gpt.py --tensor-model-parallel-size 1 --pipeline-model-parallel-size 1 --num-layers 15 --hidden-size 768 --num-attention-heads 12 --kv-channels 64 --ffn-hidden-size 3072 --seq-length 2048 --max-position-embeddings 2048 --micro-batch-size 4 --global-batch-size 256 --train-samples 5_517_578 --vocab-file gpt2/vocab.json --merge-file gpt2/merges.txt --clip-grad 1.0 --kill-switch-path kill-switch-146m14b100mdedup --bf16 --optimizer adam --adam-beta1 0.9 --adam-beta2 0.999 --adam-eps 1e-8 --lr 2e-4 --min-lr 2e-5 --lr-decay-style cosine --lr-decay-samples 5_517_578 --lr-warmup-samples 55_176 --clip-grad 1.0 --weight-decay 1e-1 --log-interval 10 --save-interval 1000 --eval-interval 1000 --eval-iters 1 --tensorboard-dir tensorboard_146m14b100mdedup --tensorboard-queue-size 5 --log-timers-to-tensorboard --log-batch-size-to-tensorboard --log-validation-ppl-to-tensorboard --save checkpoints_146m14b100mdedup --load checkpoints_146m14b100mdedup --train-weighted-split-paths-path train100mdedup.txt --valid-weighted-split-paths-path val.txt --data-impl mmap --deepspeed --deepspeed_config ds_configs/3326961.json --zero-stage 0 START 3326961: Thu 16 Mar 2023 11:53:02 PM EET 0: 0: 0: ======================= ROCm System Management Interface ======================= 0: ================================= Concise Info ================================= 0: GPU Temp AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU% 0: 0 44.0c 93.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 0: 1 50.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 0: 2 45.0c 87.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 0: 3 45.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 0: 4 42.0c 92.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 0: 5 46.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 0: 6 45.0c 91.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 0: 7 42.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 0: ================================================================================ 0: ============================= End of ROCm SMI Log ============================== 6: 6: 6: ======================= ROCm System Management Interface ======================= 6: ================================= Concise Info ================================= 6: GPU Temp AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU% 6: 0 41.0c 95.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 6: 1 46.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 6: 2 41.0c 99.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 6: 3 47.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 6: 4 46.0c 94.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 6: 5 41.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 6: 6 43.0c 94.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 6: 7 43.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 6: ================================================================================ 6: ============================= End of ROCm SMI Log ============================== 4: 4: 4: ======================= ROCm System Management Interface ======================= 4: ================================= Concise Info ================================= 4: GPU Temp AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU% 4: 0 47.0c 93.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 4: 1 43.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 4: 2 38.0c 90.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 4: 3 47.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 4: 4 37.0c 93.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 4: 5 46.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 4: 6 42.0c 83.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 4: 7 44.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 4: ================================================================================ 4: ============================= End of ROCm SMI Log ============================== 3: 3: 3: ======================= ROCm System Management Interface ======================= 3: ================================= Concise Info ================================= 3: GPU Temp AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU% 3: 0 48.0c 87.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 3: 1 46.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 3: 2 43.0c 87.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 3: 3 47.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 3: 4 41.0c 89.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 3: 5 50.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 3: 6 37.0c 91.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 3: 7 45.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 3: ================================================================================ 3: ============================= End of ROCm SMI Log ============================== 2: 2: 2: ======================= ROCm System Management Interface ======================= 2: ================================= Concise Info ================================= 2: GPU Temp AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU% 2: 0 47.0c 88.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 2: 1 48.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 2: 2 40.0c 87.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 2: 3 45.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 2: 4 46.0c 87.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 2: 5 41.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 2: 6 40.0c 90.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 2: 7 41.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 2: ================================================================================ 2: ============================= End of ROCm SMI Log ============================== 5: 5: 5: ======================= ROCm System Management Interface ======================= 5: ================================= Concise Info ================================= 5: GPU Temp AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU% 5: 0 46.0c 91.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 5: 1 46.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 5: 2 40.0c 95.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 5: 3 45.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 5: 4 42.0c 85.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 5: 5 49.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 5: 6 40.0c 93.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 5: 7 43.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 5: ================================================================================ 5: ============================= End of ROCm SMI Log ============================== 7: 7: 7: ======================= ROCm System Management Interface ======================= 7: ================================= Concise Info ================================= 7: GPU Temp AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU% 7: 0 47.0c 90.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 7: 1 45.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 7: 2 43.0c 99.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 7: 3 45.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 7: 4 44.0c 90.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 7: 5 49.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 7: 6 41.0c 88.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 7: 7 48.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 7: ================================================================================ 7: ============================= End of ROCm SMI Log ============================== 1: 1: 1: ======================= ROCm System Management Interface ======================= 1: ================================= Concise Info ================================= 1: GPU Temp AvgPwr SCLK MCLK Fan Perf PwrCap VRAM% GPU% 1: 0 42.0c 92.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 1: 1 42.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 1: 2 41.0c 82.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 1: 3 46.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 1: 4 41.0c 87.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 1: 5 46.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 1: 6 38.0c 83.0W 800Mhz 1600Mhz 0% auto 560.0W 0% 0% 1: 7 44.0c N/A 800Mhz 1600Mhz 0% auto 0.0W 0% 0% 1: ================================================================================ 1: ============================= End of ROCm SMI Log ============================== 4: Launching on nid005158 (4/8), master nid005154 port 9999, GPUs 8, CUDA: True 5: Launching on nid005159 (5/8), master nid005154 port 9999, GPUs 8, CUDA: True 3: Launching on nid005157 (3/8), master nid005154 port 9999, GPUs 8, CUDA: True 2: Launching on nid005156 (2/8), master nid005154 port 9999, GPUs 8, CUDA: True 0: Launching on nid005154 (0/8), master nid005154 port 9999, GPUs 8, CUDA: True 1: Launching on nid005155 (1/8), master nid005154 port 9999, GPUs 8, CUDA: True 7: Launching on nid005161 (7/8), master nid005154 port 9999, GPUs 8, CUDA: True 6: Launching on nid005160 (6/8), master nid005154 port 9999, GPUs 8, CUDA: True 0: using world size: 64, data-parallel-size: 64, tensor-model-parallel size: 1, pipeline-model-parallel size: 1 0: accumulate and all-reduce gradients in fp32 for bfloat16 data type. 0: using torch.bfloat16 for parameters ... 0: ------------------------ arguments ------------------------ 0: abort_on_unmet_fused_kernel_constraints ......... False 0: accumulate_allreduce_grads_in_fp32 .............. True 0: adam_beta1 ...................................... 0.9 0: adam_beta2 ...................................... 0.999 0: adam_eps ........................................ 1e-08 0: adlr_autoresume ................................. False 0: adlr_autoresume_interval ........................ 1000 0: apply_query_key_layer_scaling ................... True 0: apply_residual_connection_post_layernorm ........ False 0: attention_dropout ............................... 0.1 0: attention_softmax_in_fp32 ....................... False 0: bert_binary_head ................................ True 0: bert_load ....................................... None 0: bf16 ............................................ True 0: bias_dropout_fusion ............................. True 0: bias_gelu_fusion ................................ True 0: biencoder_projection_dim ........................ 0 0: biencoder_shared_query_context_model ............ False 0: block_data_path ................................. None 0: checkpoint_activations .......................... False 0: checkpoint_in_cpu ............................... False 0: checkpoint_num_layers ........................... 1 0: clip_grad ....................................... 1.0 0: codecarbon_dir .................................. None 0: consumed_train_samples .......................... 0 0: consumed_train_tokens ........................... 0 0: consumed_valid_samples .......................... 0 0: contigious_checkpointing ........................ False 0: cpu_optimizer ................................... False 0: cpu_torch_adam .................................. False 0: curriculum_learning ............................. False 0: data_impl ....................................... mmap 0: data_parallel_size .............................. 64 0: data_path ....................................... None 0: dataloader_type ................................. single 0: DDP_impl ........................................ local 0: decoder_seq_length .............................. None 0: deepscale ....................................... False 0: deepscale_config ................................ None 0: deepspeed ....................................... True 0: deepspeed_activation_checkpointing .............. False 0: deepspeed_config ................................ ds_configs/3326961.json 0: deepspeed_mpi ................................... False 0: distribute_checkpointed_activations ............. False 0: distributed_backend ............................. nccl 0: embed_layernorm ................................. False 0: embedding_path .................................. None 0: encoder_seq_length .............................. 2048 0: eod_mask_loss ................................... False 0: eval_interval ................................... 1000 0: eval_iters ...................................... 1 0: eval_only ....................................... None 0: evidence_data_path .............................. None 0: exit_duration_in_mins ........................... None 0: exit_interval ................................... None 0: ffn_hidden_size ................................. 3072 0: finetune ........................................ False 0: fp16 ............................................ False 0: fp16_lm_cross_entropy ........................... False 0: fp32_residual_connection ........................ False 0: gigaflos_no_embeds .............................. 0 0: global_batch_size ............................... 256 0: glu_activation .................................. None 0: hidden_dropout .................................. 0.1 0: hidden_size ..................................... 768 0: hysteresis ...................................... 2 0: ict_head_size ................................... None 0: ict_load ........................................ None 0: img_dim ......................................... 224 0: indexer_batch_size .............................. 128 0: indexer_log_interval ............................ 1000 0: inference ....................................... False 0: init_method_std ................................. 0.02 0: init_method_xavier_uniform ...................... False 0: initial_loss_scale .............................. 4294967296 0: kill_switch_path ................................ kill-switch-146m14b100mdedup 0: kv_channels ..................................... 64 0: layer_norm_fusion ............................... True 0: layernorm_epsilon ............................... 1e-05 0: lazy_mpu_init ................................... None 0: load ............................................ checkpoints_146m14b100mdedup 0: local_rank ...................................... None 0: log_batch_size_to_tensorboard ................... True 0: log_interval .................................... 10 0: log_learning_rate_to_tensorboard ................ True 0: log_level ....................................... None 0: log_level_replica ............................... None 0: log_loss_scale_to_tensorboard ................... True 0: log_num_zeros_in_grad ........................... False 0: log_params_norm ................................. False 0: log_path ........................................ None 0: log_timers_to_tensorboard ....................... True 0: log_validation_ppl_to_tensorboard ............... True 0: loss_on_targets_only ............................ False 0: loss_scale ...................................... None 0: loss_scale_window ............................... 1000 0: lr .............................................. 0.0002 0: lr_decay_iters .................................. None 0: lr_decay_samples ................................ 5517578 0: lr_decay_style .................................. cosine 0: lr_decay_tokens ................................. None 0: lr_warmup_fraction .............................. None 0: lr_warmup_iters ................................. 0 0: lr_warmup_samples ............................... 55176 0: make_vocab_size_divisible_by .................... 128 0: mask_prob ....................................... 0.15 0: masked_softmax_fusion ........................... True 0: max_position_embeddings ......................... 2048 0: mean_noise_span_length .......................... None 0: memory_centric_tiled_linear ..................... False 0: merge_file ...................................... gpt2/merges.txt 0: micro_batch_size ................................ 4 0: min_loss_scale .................................. 1.0 0: min_lr .......................................... 2e-05 0: mmap_warmup ..................................... False 0: no_load_optim ................................... None 0: no_load_rng ..................................... None 0: no_save_optim ................................... None 0: no_save_rng ..................................... None 0: noise_density ................................... None 0: num_attention_heads ............................. 12 0: num_channels .................................... 3 0: num_classes ..................................... 1000 0: num_layers ...................................... 15 0: num_layers_per_virtual_pipeline_stage ........... None 0: num_workers ..................................... 2 0: onnx_safe ....................................... None 0: openai_gelu ..................................... False 0: optimizer ....................................... adam 0: optimizer_fusion ................................ True 0: override_lr_scheduler ........................... False 0: pad_vocab_size_to ............................... None 0: params_dtype .................................... torch.bfloat16 0: partition_activations ........................... False 0: patch_dim ....................................... 16 0: pipeline_model_parallel_size .................... 1 0: position_embedding_type ......................... PositionEmbeddingType.absolute 0: pp_partition_method ............................. None 0: profile_backward ................................ False 0: query_in_block_prob ............................. 0.1 0: rampup_batch_size ............................... None 0: rank ............................................ 0 0: remote_device ................................... none 0: reset_attention_mask ............................ False 0: reset_position_ids .............................. False 0: reset_progress .................................. None 0: retriever_report_topk_accuracies ................ [] 0: retriever_score_scaling ......................... False 0: retriever_seq_length ............................ 256 0: reweight_loss_based_on_position_frequency ....... False 0: sample_rate ..................................... 1.0 0: save ............................................ checkpoints_146m14b100mdedup 0: save_interval ................................... 1000 0: scatter_gather_tensors_in_pipeline .............. True 0: scattered_embeddings ............................ False 0: seed ............................................ 1234 0: seq_length ...................................... 2048 0: sgd_momentum .................................... 0.9 0: short_seq_prob .................................. 0.1 0: skip_train_iteration_range ...................... None 0: split ........................................... None 0: split_transformers .............................. False 0: sync_tp_duplicated_parameters ................... False 0: synchronize_each_layer .......................... False 0: tensor_model_parallel_size ...................... 1 0: tensorboard_dir ................................. tensorboard_146m14b100mdedup 0: tensorboard_log_interval ........................ 1 0: tensorboard_queue_size .......................... 5 0: test_weighted_split_paths ....................... None 0: test_weighted_split_paths_path .................. None 0: tile_factor ..................................... 1 0: titles_data_path ................................ None 0: tokenizer_name_or_path .......................... None 0: tokenizer_type .................................. GPT2BPETokenizer 0: train_iters ..................................... None 0: train_samples ................................... 5517578 0: train_tokens .................................... None 0: train_weighted_split_names ...................... ['train'] 0: train_weighted_split_paths ...................... [['/scratch/project_462000119/data/c4_subsampled/gpt2tok_c4_en_dedup_100M_text_document']] 0: train_weighted_split_paths_path ................. None 0: train_weighted_split_splits ..................... [['0:1']] 0: train_weighted_split_weights .................... [['1.0']] 0: universal_checkpoint ............................ False 0: use_bnb_optimizer ............................... False 0: use_checkpoint_lr_scheduler ..................... False 0: use_contiguous_buffers_in_ddp ................... True 0: use_cpu_initialization .......................... None 0: use_one_sent_docs ............................... False 0: use_pin_memory .................................. False 0: valid_num_workers ............................... 2 0: valid_weighted_split_names ...................... ['validation'] 0: valid_weighted_split_paths ...................... [['/scratch/project_462000119/data/c4_validation/gpt2tok_c4validation_rerun_text_document']] 0: valid_weighted_split_paths_path ................. None 0: valid_weighted_split_splits ..................... [['0:1']] 0: valid_weighted_split_weights .................... [['1.0']] 0: virtual_pipeline_model_parallel_size ............ None 0: vocab_extra_ids ................................. 0 0: vocab_file ...................................... gpt2/vocab.json 0: weight_decay .................................... 0.1 0: world_size ...................................... 64 0: zero_allgather_bucket_size ...................... 0.0 0: zero_contigious_gradients ....................... False 0: zero_reduce_bucket_size ......................... 0.0 0: zero_reduce_scatter ............................. False 0: zero_stage ...................................... 0 0: -------------------- end of arguments --------------------- 0: setting number of micro-batches to constant 1 0: > building GPT2BPETokenizer tokenizer ... 0: > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304) 0: DeepSpeed general environment info: 0: torch install path ............... ['/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch'] 0: torch version .................... 1.13.0+rocm5.2 0: torch cuda version ............... None 0: torch hip version ................ 5.2.21151-afdc89f8 0: nvcc version ..................... None 0: deepspeed install path ........... ['/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed'] 0: deepspeed info ................... 0.7.5, unknown, unknown 0: deepspeed wheel compiled w. ...... torch 1.13, hip 5.1 7: > setting tensorboard ... 0: **** Git info for Megatron: git_hash=unknown git_branch=unknown **** 0: > initializing torch distributed ... 0: [2023-03-16 23:54:34,672] [INFO] [comm.py:633:init_distributed] Initializing TorchBackend in DeepSpeed with backend nccl 0: > initializing tensor model parallel with size 1 0: > initializing pipeline model parallel with size 1 0: > setting random seeds to 1234 ... 0: > initializing model parallel cuda seeds on global rank 0, model parallel rank 0, and data parallel rank 0 with model parallel seed: 3952 and data parallel seed: 1234 0: > compiling dataset index builder ... 0: make: Entering directory '/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/data' 0: make: Nothing to be done for 'default'. 0: make: Leaving directory '/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/data' 0: >>> done with dataset index builder. Compilation time: 0.112 seconds 0: > compiling and loading fused kernels ... 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax.cpp -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax_hip.cpp [skipped, already hipified] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax_hip.h [skipped, already hipified] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/compat.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/compat.h [skipped, no changes] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/type_shim.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/type_shim.h [skipped, no changes] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax_cuda.cu -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax_hip.hip [skipped, already hipified] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/type_shim.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/type_shim.h [skipped, no changes] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/compat.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/compat.h [skipped, no changes] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_upper_triang_masked_softmax.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_upper_triang_masked_softmax_hip.h [skipped, already hipified] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax_hip.h [skipped, already hipified] 0: Total number of unsupported CUDA function calls: 0 0: 0: 0: Total number of replaced kernel launches: 102 0: ninja: no work to do. 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/layer_norm_cuda.cpp -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/layer_norm_cuda.cpp [skipped, no changes] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/layer_norm_cuda_kernel.cu -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/layer_norm_hip_kernel.hip [skipped, already hipified] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/type_shim.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/type_shim.h [skipped, no changes] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/compat.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/compat.h [skipped, no changes] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_upper_triang_masked_softmax.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_upper_triang_masked_softmax_hip.h [skipped, already hipified] 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax.h -> /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/scaled_masked_softmax_hip.h [skipped, already hipified] 0: Total number of unsupported CUDA function calls: 0 0: 0: 0: Total number of replaced kernel launches: 67 0: ninja: no work to do. 0: >>> done with compiling and loading fused kernels. Compilation time: 29.351 seconds 0: time to initialize megatron (seconds): 58.816 0: [after megatron is initialized] datetime: 2023-03-16 23:55:06 0: building GPT model ... 0: [2023-03-16 23:55:06,948] [INFO] [utils.py:827:see_memory_usage] Before Building Model 0: [2023-03-16 23:55:06,949] [INFO] [utils.py:828:see_memory_usage] MA 0.0 GB Max_MA 0.0 GB CA 0.0 GB Max_CA 0 GB 0: [2023-03-16 23:55:06,949] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 33.19 GB, percent = 6.6% 0: SEED_LAYERS=False BASE_SEED=1234 SEED_FN=None 0: Using topology: {ProcessCoord(pipe=0, data=0, model=0): 0, ProcessCoord(pipe=0, data=1, model=0): 1, ProcessCoord(pipe=0, data=2, model=0): 2, ProcessCoord(pipe=0, data=3, model=0): 3, ProcessCoord(pipe=0, data=4, model=0): 4, ProcessCoord(pipe=0, data=5, model=0): 5, ProcessCoord(pipe=0, data=6, model=0): 6, ProcessCoord(pipe=0, data=7, model=0): 7, ProcessCoord(pipe=0, data=8, model=0): 8, ProcessCoord(pipe=0, data=9, model=0): 9, ProcessCoord(pipe=0, data=10, model=0): 10, ProcessCoord(pipe=0, data=11, model=0): 11, ProcessCoord(pipe=0, data=12, model=0): 12, ProcessCoord(pipe=0, data=13, model=0): 13, ProcessCoord(pipe=0, data=14, model=0): 14, ProcessCoord(pipe=0, data=15, model=0): 15, ProcessCoord(pipe=0, data=16, model=0): 16, ProcessCoord(pipe=0, data=17, model=0): 17, ProcessCoord(pipe=0, data=18, model=0): 18, ProcessCoord(pipe=0, data=19, model=0): 19, ProcessCoord(pipe=0, data=20, model=0): 20, ProcessCoord(pipe=0, data=21, model=0): 21, ProcessCoord(pipe=0, data=22, model=0): 22, ProcessCoord(pi 0: pe=0, data=23, model=0): 23, ProcessCoord(pipe=0, data=24, model=0): 24, ProcessCoord(pipe=0, data=25, model=0): 25, ProcessCoord(pipe=0, data=26, model=0): 26, ProcessCoord(pipe=0, data=27, model=0): 27, ProcessCoord(pipe=0, data=28, model=0): 28, ProcessCoord(pipe=0, data=29, model=0): 29, ProcessCoord(pipe=0, data=30, model=0): 30, ProcessCoord(pipe=0, data=31, model=0): 31, ProcessCoord(pipe=0, data=32, model=0): 32, ProcessCoord(pipe=0, data=33, model=0): 33, ProcessCoord(pipe=0, data=34, model=0): 34, ProcessCoord(pipe=0, data=35, model=0): 35, ProcessCoord(pipe=0, data=36, model=0): 36, ProcessCoord(pipe=0, data=37, model=0): 37, ProcessCoord(pipe=0, data=38, model=0): 38, ProcessCoord(pipe=0, data=39, model=0): 39, ProcessCoord(pipe=0, data=40, model=0): 40, ProcessCoord(pipe=0, data=41, model=0): 41, ProcessCoord(pipe=0, data=42, model=0): 42, ProcessCoord(pipe=0, data=43, model=0): 43, ProcessCoord(pipe=0, data=44, model=0): 44, ProcessCoord(pipe=0, data=45, model=0): 45, ProcessCoord(pipe=0, data=4 0: 6, model=0): 46, ProcessCoord(pipe=0, data=47, model=0): 47, ProcessCoord(pipe=0, data=48, model=0): 48, ProcessCoord(pipe=0, data=49, model=0): 49, ProcessCoord(pipe=0, data=50, model=0): 50, ProcessCoord(pipe=0, data=51, model=0): 51, ProcessCoord(pipe=0, data=52, model=0): 52, ProcessCoord(pipe=0, data=53, model=0): 53, ProcessCoord(pipe=0, data=54, model=0): 54, ProcessCoord(pipe=0, data=55, model=0): 55, ProcessCoord(pipe=0, data=56, model=0): 56, ProcessCoord(pipe=0, data=57, model=0): 57, ProcessCoord(pipe=0, data=58, model=0): 58, ProcessCoord(pipe=0, data=59, model=0): 59, ProcessCoord(pipe=0, data=60, model=0): 60, ProcessCoord(pipe=0, data=61, model=0): 61, ProcessCoord(pipe=0, data=62, model=0): 62, ProcessCoord(pipe=0, data=63, model=0): 63} 0: [2023-03-16 23:55:08,955] [INFO] [module.py:366:_partition_layers] Partitioning pipeline stages with method type:transformer 0: stage=0 layers=22 0: 0: _to_float16 0: 1: EmbeddingPipe 0: 2: 0: 3: ParallelTransformerLayerPipe 0: 4: ParallelTransformerLayerPipe 0: 5: ParallelTransformerLayerPipe 0: 6: ParallelTransformerLayerPipe 0: 7: ParallelTransformerLayerPipe 0: 8: ParallelTransformerLayerPipe 0: 9: ParallelTransformerLayerPipe 0: 10: ParallelTransformerLayerPipe 0: 11: ParallelTransformerLayerPipe 0: 12: ParallelTransformerLayerPipe 0: 13: ParallelTransformerLayerPipe 0: 14: ParallelTransformerLayerPipe 0: 15: ParallelTransformerLayerPipe 0: 16: ParallelTransformerLayerPipe 0: 17: ParallelTransformerLayerPipe 0: 18: undo 0: 19: MixedFusedLayerNorm 0: 20: EmbeddingPipe 0: 21: float16_to_fp32 0: loss: CrossEntropy 0: [2023-03-16 23:55:09,214] [INFO] [utils.py:827:see_memory_usage] After Building Model 0: [2023-03-16 23:55:09,214] [INFO] [utils.py:828:see_memory_usage] MA 0.28 GB Max_MA 0.28 GB CA 0.29 GB Max_CA 0 GB 0: [2023-03-16 23:55:09,214] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 33.21 GB, percent = 6.6% 0: setting training iterations to 21553 0: > learning rate decay style: cosine 0: DeepSpeed is enabled. 0: [2023-03-16 23:55:09,216] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed info: version=0.7.5, git-hash=unknown, git-branch=unknown 0: [2023-03-16 23:55:22,036] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed Flops Profiler Enabled: False 0: [2023-03-16 23:55:22,036] [INFO] [logging.py:68:log_dist] [Rank 0] Removing param_group that has no 'params' in the client Optimizer 0: [2023-03-16 23:55:22,036] [INFO] [logging.py:68:log_dist] [Rank 0] Using client Optimizer as basic optimizer 0: [2023-03-16 23:55:22,041] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed Basic Optimizer = FusedAdam 0: [2023-03-16 23:55:22,041] [INFO] [logging.py:68:log_dist] [Rank 0] Creating BF16 optimizer 0: [2023-03-16 23:55:22,159] [INFO] [utils.py:827:see_memory_usage] begin bf16_optimizer 0: [2023-03-16 23:55:22,159] [INFO] [utils.py:828:see_memory_usage] MA 0.28 GB Max_MA 0.29 GB CA 0.31 GB Max_CA 0 GB 0: [2023-03-16 23:55:22,160] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 33.89 GB, percent = 6.7% 0: ninja: no work to do. 0: Time to load utils op: 0.11079978942871094 seconds 0: Time to load utils op: 0.229996919631958 seconds 3: Time to load utils op: 0.2296130657196045 seconds 1: Time to load utils op: 0.23061680793762207 seconds 0: [2023-03-16 23:55:22,399] [INFO] [utils.py:827:see_memory_usage] before initializing group 0 0: [2023-03-16 23:55:22,399] [INFO] [utils.py:828:see_memory_usage] MA 0.28 GB Max_MA 0.28 GB CA 0.31 GB Max_CA 0 GB 0: [2023-03-16 23:55:22,399] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 33.9 GB, percent = 6.7% 0: ninja: no work to do. 0: Time to load utils op: 0.0006780624389648438 seconds 3: Time to load utils op: 0.0004932880401611328 seconds 0: Time to load utils op: 0.1432628631591797 seconds 1: Time to load utils op: 0.0005805492401123047 seconds 0: Time to load utils op: 0.00045990943908691406 seconds 0: Time to load utils op: 0.20272254943847656 seconds 0: Time to load utils op: 0.2027428150177002 secondsTime to load utils op: 0.20261836051940918 seconds 0: 0: Time to load utils op: 0.20299029350280762 secondsTime to load utils op: 0.20264673233032227 seconds 0: 1: Time to load utils op: 0.20398879051208496 secondsTime to load utils op: 0.20388507843017578 seconds 1: Time to load utils op: 0.20404410362243652 seconds 1: 1: Time to load utils op: 0.20357370376586914 seconds 3: Time to load utils op: 0.20411014556884766 secondsTime to load utils op: 0.20410656929016113 seconds 3: 3: Time to load utils op: 0.20415592193603516 seconds 3: Time to load utils op: 0.2043626308441162 seconds 3: Time to load utils op: 0.20416545867919922 seconds 3: Time to load utils op: 0.20446085929870605 secondsTime to load utils op: 0.20443201065063477 seconds 3: 1: Time to load utils op: 0.2021634578704834 seconds 1: Time to load utils op: 0.20222997665405273 seconds 1: Time to load utils op: 0.20242762565612793 seconds 0: Time to load utils op: 0.00035381317138671875 seconds 0: Time to load utils op: 0.0003936290740966797 secondsTime to load utils op: 0.0004405975341796875 seconds 0: 0: Time to load utils op: 0.0004031658172607422 seconds 0: Time to load utils op: 0.00039577484130859375 seconds 0: [2023-03-16 23:55:22,527] [INFO] [utils.py:827:see_memory_usage] after initializing group 0 0: [2023-03-16 23:55:22,528] [INFO] [utils.py:828:see_memory_usage] MA 0.62 GB Max_MA 0.62 GB CA 0.82 GB Max_CA 1 GB 0: [2023-03-16 23:55:22,528] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 33.95 GB, percent = 6.7% 3: Time to load utils op: 0.0003116130828857422 secondsTime to load utils op: 0.0003254413604736328 seconds 3: 3: Time to load utils op: 0.0003948211669921875 seconds 3: Time to load utils op: 0.0003590583801269531 seconds 3: Time to load utils op: 0.0003666877746582031 seconds 3: Time to load utils op: 0.00036644935607910156 seconds 3: Time to load utils op: 0.0003612041473388672 seconds 1: Time to load utils op: 0.0004146099090576172 seconds 1: Time to load utils op: 0.00039076805114746094 seconds 1: Time to load utils op: 0.0003838539123535156 seconds 1: Time to load utils op: 0.00037360191345214844 seconds 4: Time to load utils op: 0.21976900100708008 seconds 4: Time to load utils op: 0.2201247215270996 secondsTime to load utils op: 0.22243785858154297 seconds 4: 4: Time to load utils op: 0.22034120559692383 secondsTime to load utils op: 0.2189033031463623 seconds 4: 1: Time to load utils op: 0.0004107952117919922 seconds 4: Time to load utils op: 0.21880340576171875 seconds 4: Time to load utils op: 0.22030353546142578 seconds 4: Time to load utils op: 0.2174520492553711 seconds 1: Time to load utils op: 0.00033354759216308594 seconds 1: Time to load utils op: 0.0003342628479003906 seconds 2: Time to load utils op: 0.2209916114807129 seconds 2: Time to load utils op: 0.22103667259216309 seconds 2: Time to load utils op: 0.22105789184570312 seconds 2: Time to load utils op: 0.22108101844787598 secondsTime to load utils op: 0.22107148170471191 seconds 2: Time to load utils op: 0.22107601165771484 seconds 2: 2: Time to load utils op: 0.22108149528503418 seconds 2: Time to load utils op: 0.221083402633667 seconds 6: Time to load utils op: 0.22038698196411133 secondsTime to load utils op: 0.21965408325195312 seconds 6: 6: Time to load utils op: 0.22031736373901367 seconds 6: Time to load utils op: 0.22047758102416992 seconds 6: Time to load utils op: 0.2200336456298828 secondsTime to load utils op: 0.22098398208618164 seconds 6: 6: Time to load utils op: 0.21962404251098633 seconds 6: Time to load utils op: 0.22015643119812012 seconds 7: Time to load utils op: 0.21971678733825684 seconds 7: Time to load utils op: 0.21972346305847168 seconds 7: Time to load utils op: 0.22008585929870605 seconds 7: Time to load utils op: 0.2204282283782959 seconds 7: Time to load utils op: 0.22023653984069824 seconds 7: Time to load utils op: 0.21975970268249512 seconds 7: Time to load utils op: 0.2201075553894043 seconds 7: Time to load utils op: 0.21979331970214844 seconds 5: Time to load utils op: 0.2192525863647461 seconds 5: Time to load utils op: 0.2192537784576416 seconds 5: Time to load utils op: 0.2192363739013672 seconds 5: Time to load utils op: 0.21927118301391602 seconds 5: Time to load utils op: 0.21927475929260254 seconds 5: Time to load utils op: 0.219268798828125 secondsTime to load utils op: 0.219285249710083 seconds 5: 5: Time to load utils op: 0.21928954124450684 seconds 4: Time to load utils op: 0.0005037784576416016 seconds 4: Time to load utils op: 0.0005273818969726562 seconds 4: Time to load utils op: 0.00055694580078125 seconds 4: Time to load utils op: 0.0005178451538085938 seconds 4: Time to load utils op: 0.0005648136138916016 seconds 4: Time to load utils op: 0.0006315708160400391 seconds 4: Time to load utils op: 0.0005965232849121094 seconds 4: Time to load utils op: 0.0006680488586425781 seconds 6: Time to load utils op: 0.0009503364562988281 seconds 6: Time to load utils op: 0.0011467933654785156 seconds 6: Time to load utils op: 0.0011494159698486328 seconds 6: Time to load utils op: 0.001344442367553711 seconds 6: Time to load utils op: 0.0012180805206298828 secondsTime to load utils op: 0.0012958049774169922 seconds 6: 6: Time to load utils op: 0.0013527870178222656 seconds 6: Time to load utils op: 0.0013587474822998047 seconds 7: Time to load utils op: 0.0009624958038330078 secondsTime to load utils op: 0.0009865760803222656 seconds 7: 7: Time to load utils op: 0.0012221336364746094 seconds 7: Time to load utils op: 0.0013577938079833984 seconds 7: Time to load utils op: 0.0013356208801269531 secondsTime to load utils op: 0.0013496875762939453 seconds 7: Time to load utils op: 0.0013167858123779297 seconds 7: 7: Time to load utils op: 0.0013942718505859375 seconds 5: Time to load utils op: 0.0008656978607177734 secondsTime to load utils op: 0.0008261203765869141 seconds 5: 5: Time to load utils op: 0.000997304916381836 secondsTime to load utils op: 0.0010046958923339844 seconds 5: 5: Time to load utils op: 0.0009815692901611328 seconds 5: Time to load utils op: 0.0010251998901367188 seconds 5: Time to load utils op: 0.0010478496551513672 seconds 5: Time to load utils op: 0.0010366439819335938 seconds 2: Time to load utils op: 0.0007462501525878906 seconds 2: Time to load utils op: 0.0010833740234375 secondsTime to load utils op: 0.0011017322540283203 seconds 2: 2: Time to load utils op: 0.0012485980987548828 seconds 2: Time to load utils op: 0.0011055469512939453 seconds 2: Time to load utils op: 0.001195669174194336 seconds 2: Time to load utils op: 0.0011925697326660156 seconds 2: Time to load utils op: 0.0011227130889892578 seconds 0: [2023-03-16 23:55:22,650] [INFO] [utils.py:827:see_memory_usage] before initializing group 1 0: [2023-03-16 23:55:22,651] [INFO] [utils.py:828:see_memory_usage] MA 0.62 GB Max_MA 0.62 GB CA 0.82 GB Max_CA 1 GB 0: [2023-03-16 23:55:22,651] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 34.04 GB, percent = 6.8% 0: [2023-03-16 23:55:22,757] [INFO] [utils.py:827:see_memory_usage] after initializing group 1 0: [2023-03-16 23:55:22,758] [INFO] [utils.py:828:see_memory_usage] MA 0.83 GB Max_MA 0.83 GB CA 1.13 GB Max_CA 1 GB 0: [2023-03-16 23:55:22,758] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 34.04 GB, percent = 6.8% 0: [2023-03-16 23:55:22,860] [INFO] [utils.py:827:see_memory_usage] before initializing group 2 0: [2023-03-16 23:55:22,860] [INFO] [utils.py:828:see_memory_usage] MA 0.83 GB Max_MA 0.83 GB CA 1.13 GB Max_CA 1 GB 0: [2023-03-16 23:55:22,860] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 34.04 GB, percent = 6.8% 0: [2023-03-16 23:55:22,964] [INFO] [utils.py:827:see_memory_usage] after initializing group 2 0: [2023-03-16 23:55:22,965] [INFO] [utils.py:828:see_memory_usage] MA 0.83 GB Max_MA 0.83 GB CA 1.13 GB Max_CA 1 GB 0: [2023-03-16 23:55:22,965] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 34.04 GB, percent = 6.8% 0: [2023-03-16 23:55:23,066] [INFO] [utils.py:827:see_memory_usage] before initialize_optimizer 0: [2023-03-16 23:55:23,067] [INFO] [utils.py:828:see_memory_usage] MA 0.83 GB Max_MA 0.83 GB CA 1.13 GB Max_CA 1 GB 0: [2023-03-16 23:55:23,067] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 34.04 GB, percent = 6.8% 0: [2023-03-16 23:55:23,174] [INFO] [utils.py:827:see_memory_usage] end initialize_optimizer 0: [2023-03-16 23:55:23,174] [INFO] [utils.py:828:see_memory_usage] MA 0.85 GB Max_MA 0.85 GB CA 1.13 GB Max_CA 1 GB 0: [2023-03-16 23:55:23,174] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 34.04 GB, percent = 6.8% 0: [2023-03-16 23:55:23,275] [INFO] [utils.py:827:see_memory_usage] end bf16_optimizer 0: [2023-03-16 23:55:23,275] [INFO] [utils.py:828:see_memory_usage] MA 0.85 GB Max_MA 0.85 GB CA 1.13 GB Max_CA 1 GB 0: [2023-03-16 23:55:23,276] [INFO] [utils.py:836:see_memory_usage] CPU Virtual Memory: used = 34.04 GB, percent = 6.8% 0: [2023-03-16 23:55:23,276] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed Final Optimizer = FusedAdam 0: [2023-03-16 23:55:23,276] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed using client LR scheduler 0: [2023-03-16 23:55:23,276] [INFO] [logging.py:68:log_dist] [Rank 0] DeepSpeed LR Scheduler = 0: [2023-03-16 23:55:23,276] [INFO] [logging.py:68:log_dist] [Rank 0] step=0, skipped=0, lr=[0.0, 0.0, 0.0], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 0: [2023-03-16 23:55:23,276] [INFO] [config.py:1007:print] DeepSpeedEngine configuration: 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] activation_checkpointing_config { 0: "partition_activations": false, 0: "contiguous_memory_optimization": false, 0: "cpu_checkpointing": false, 0: "number_checkpoints": null, 0: "synchronize_checkpoint_boundary": false, 0: "profile": false 0: } 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] aio_config ................... {'block_size': 1048576, 'queue_depth': 8, 'thread_count': 1, 'single_submit': False, 'overlap_events': True} 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] amp_enabled .................. False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] amp_params ................... False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] autotuning_config ............ { 0: "enabled": false, 0: "start_step": null, 0: "end_step": null, 0: "metric_path": null, 0: "arg_mappings": null, 0: "metric": "throughput", 0: "model_info": null, 0: "results_dir": "/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/autotuning_results", 0: "exps_dir": "/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/autotuning_exps", 0: "overwrite": true, 0: "fast": true, 0: "start_profile_step": 3, 0: "end_profile_step": 5, 0: "tuner_type": "gridsearch", 0: "tuner_early_stopping": 5, 0: "tuner_num_trials": 50, 0: "model_info_path": null, 0: "mp_size": 1, 0: "max_train_batch_size": null, 0: "min_train_batch_size": 1, 0: "max_train_micro_batch_size_per_gpu": 1.024000e+03, 0: "min_train_micro_batch_size_per_gpu": 1, 0: "num_tuning_micro_batch_sizes": 3 0: } 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] bfloat16_enabled ............. True 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] checkpoint_parallel_write_pipeline False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] checkpoint_tag_validation_enabled True 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] checkpoint_tag_validation_fail False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] comms_config ................. 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] communication_data_type ...... None 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] compression_config ........... {'weight_quantization': {'shared_parameters': {'enabled': False, 'quantizer_kernel': False, 'schedule_offset': 0, 'quantize_groups': 1, 'quantize_verbose': False, 'quantization_type': 'symmetric', 'quantize_weight_in_forward': False, 'rounding': 'nearest', 'fp16_mixed_quantize': False, 'quantize_change_ratio': 0.001}, 'different_groups': {}}, 'activation_quantization': {'shared_parameters': {'enabled': False, 'quantization_type': 'symmetric', 'range_calibration': 'dynamic', 'schedule_offset': 1000}, 'different_groups': {}}, 'sparse_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'row_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'head_pruning': {'shared_parameters': {'enabled': False, 'method': 'topk', 'schedule_offset': 1000}, 'different_groups': {}}, 'channel_pruning': {'shared_pa 0: rameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'layer_reduction': {'enabled': False}} 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] curriculum_enabled ........... False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] curriculum_params ............ False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] dataloader_drop_last ......... False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] disable_allgather ............ False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] dump_state ................... False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] dynamic_loss_scale_args ...... None 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] eigenvalue_enabled ........... False 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] eigenvalue_gas_boundary_resolution 1 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] eigenvalue_layer_name ........ bert.encoder.layer 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] eigenvalue_layer_num ......... 0 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] eigenvalue_max_iter .......... 100 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] eigenvalue_stability ......... 1e-06 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] eigenvalue_tol ............... 0.01 0: [2023-03-16 23:55:23,277] [INFO] [config.py:1011:print] eigenvalue_verbose ........... False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] elasticity_enabled ........... False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] flops_profiler_config ........ { 0: "enabled": false, 0: "profile_step": 1, 0: "module_depth": -1, 0: "top_modules": 1, 0: "detailed": true, 0: "output_file": null 0: } 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] fp16_auto_cast ............... None 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] fp16_enabled ................. False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] fp16_master_weights_and_gradients False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] global_rank .................. 0 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] gradient_accumulation_steps .. 1 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] gradient_clipping ............ 1.0 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] gradient_predivide_factor .... 1.0 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] initial_dynamic_scale ........ 1 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] load_universal_checkpoint .... False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] loss_scale ................... 1.0 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] memory_breakdown ............. False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] monitor_config ............... 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] nebula_config ................ { 0: "enabled": false, 0: "persistent_storage_path": null, 0: "persistent_time_interval": 100, 0: "num_of_version_in_retention": 2, 0: "enable_nebula_load": true, 0: "load_path": null 0: } 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] optimizer_legacy_fusion ...... False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] optimizer_name ............... None 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] optimizer_params ............. None 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0} 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] pld_enabled .................. False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] pld_params ................... False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] prescale_gradients ........... False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] scheduler_name ............... None 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] scheduler_params ............. None 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] sparse_attention ............. None 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] sparse_gradients_enabled ..... False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] steps_per_print .............. 2000 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] train_batch_size ............. 256 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] train_micro_batch_size_per_gpu 4 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] use_node_local_storage ....... False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] wall_clock_breakdown ......... False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] world_size ................... 64 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] zero_allow_untested_optimizer False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] zero_config .................. stage=0 contiguous_gradients=True reduce_scatter=True reduce_bucket_size=500000000 allgather_partitions=True allgather_bucket_size=500000000 overlap_comm=False load_from_fp32_weights=True elastic_checkpoint=False offload_param=None offload_optimizer=None sub_group_size=1000000000 cpu_offload_param=None cpu_offload_use_pin_memory=None cpu_offload=None prefetch_bucket_size=50000000 param_persistence_threshold=100000 model_persistence_threshold=9223372036854775807 max_live_parameters=1000000000 max_reuse_distance=1000000000 gather_16bit_weights_on_model_save=False stage3_gather_fp16_weights_on_model_save=False ignore_unused_parameters=True legacy_stage1=False round_robin_gradients=False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] zero_enabled ................. False 0: [2023-03-16 23:55:23,278] [INFO] [config.py:1011:print] zero_optimization_stage ...... 0 0: [2023-03-16 23:55:23,278] [INFO] [config.py:996:print_user_config] json = { 0: "train_micro_batch_size_per_gpu": 4, 0: "train_batch_size": 256, 0: "gradient_clipping": 1.0, 0: "zero_optimization": { 0: "stage": 0 0: }, 0: "bf16": { 0: "enabled": true 0: }, 0: "steps_per_print": 2.000000e+03, 0: "wall_clock_breakdown": false 0: } 0: Time to load utils op: 0.0004055500030517578 seconds 0: [2023-03-16 23:55:23,279] [INFO] [engine.py:87:__init__] CONFIG: micro_batches=1 micro_batch_size=4 0: [2023-03-16 23:55:23,331] [INFO] [engine.py:145:__init__] RANK=0 STAGE=0 LAYERS=22 [0, 22) STAGE_PARAMS=146525952 (146.526M) TOTAL_PARAMS=146525952 (146.526M) UNIQUE_PARAMS=146525952 (146.526M) 0: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 6: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 4: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 5: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 0: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 6: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 2: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 0: WARNING: could not find the metadata file checkpoints_146m14b100mdedup 0: will not load any checkpoints and will start from random 4: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 3: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 7: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 6: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 5: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 2: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 4: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 4: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 3: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 6: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 0: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 5: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 7: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 0: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 3: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 6: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 2: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 7: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 4: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 7: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 2: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 5: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 6: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 4: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 2: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 0: [2023-03-16 23:55:23,339] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 3: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 5: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 7: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 3: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 2: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 0: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 1: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 6: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 7: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 5: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 3: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 0: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 4: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 7: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 3: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 2: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 1: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 0: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 5: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 4: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 6: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 2: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 3: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 7: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 1: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 1: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 5: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 1: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 1: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 1: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 1: [2023-03-16 23:55:23,340] [WARNING] [engine.py:2581:load_checkpoint] Unable to find latest file at checkpoints_146m14b100mdedup/latest, if trying to load latest checkpoint please ensure this file exists or pass an explicit checkpoint tag when loading a checkpoint. 7: time (ms) | load-checkpoint: 9.01 0: estimated model parameters: 0.146525952 0: estimated model parameters without embeddings: 0.106319616 0: [after model, optimizer, and learning rate scheduler are built] datetime: 2023-03-16 23:55:23 0: > building train, validation, and test datasets ... 0: > datasets target sizes (minimum size): 0: train: 5517578 0: validation: 5632 0: test: 256 0: > building train, validation, and test datasets for GPT ... 0: > building dataset index ... 0: reading sizes... 0: reading pointers... 0: reading document index... 0: creating numpy buffer of mmap... 0: creating memory view of numpy buffer... 0: > finished creating indexed dataset in 0.040058 seconds 0: number of documents: 409500 0: > dataset split: 0: train: 0: document indices in [0, 409500) total of 409500 documents 0: > loading doc-idx mapping from /scratch/project_462000119/data/c4_subsampled/gpt2tok_c4_en_dedup_100M_text_document_train_indexmap_5517578ns_2048sl_1234s_doc_idx.npy 0: > loading sample-idx mapping from /scratch/project_462000119/data/c4_subsampled/gpt2tok_c4_en_dedup_100M_text_document_train_indexmap_5517578ns_2048sl_1234s_sample_idx.npy 0: > loading shuffle-idx mapping from /scratch/project_462000119/data/c4_subsampled/gpt2tok_c4_en_dedup_100M_text_document_train_indexmap_5517578ns_2048sl_1234s_shuffle_idx.npy 0: loaded indexed file in 0.055 seconds 0: total number of samples: 5552393 0: total number of epochs: 115 0: > building dataset index ... 0: reading sizes... 0: reading pointers... 0: reading document index... 0: creating numpy buffer of mmap... 0: creating memory view of numpy buffer... 0: > finished creating indexed dataset in 0.023232 seconds 0: number of documents: 364608 0: > dataset split: 0: validation: 0: document indices in [0, 364608) total of 364608 documents 0: > loading doc-idx mapping from /scratch/project_462000119/data/c4_validation/gpt2tok_c4validation_rerun_text_document_validation_indexmap_5632ns_2048sl_1234s_doc_idx.npy 0: > loading sample-idx mapping from /scratch/project_462000119/data/c4_validation/gpt2tok_c4validation_rerun_text_document_validation_indexmap_5632ns_2048sl_1234s_sample_idx.npy 0: > loading shuffle-idx mapping from /scratch/project_462000119/data/c4_validation/gpt2tok_c4validation_rerun_text_document_validation_indexmap_5632ns_2048sl_1234s_shuffle_idx.npy 0: loaded indexed file in 0.076 seconds 0: total number of samples: 84978 0: total number of epochs: 1 0: > finished creating GPT datasets ... 0: [after dataloaders are built] datetime: 2023-03-16 23:55:37 0: done with setup ... 0: training ... 0: Number of parameters: [tensor rank - pipeline rank] w/ and w/o embeddings: 7: time (ms) | model-and-optimizer-setup: 16801.87 | train/valid/test-data-iterators-setup: 13360.42 0: [000-000] 0.1465B / 0.1063B 0: [before the start of training step] datetime: 2023-03-16 23:55:37 0: [Rank 0] (after 10 iterations) memory (MB) | allocated: 2734.17236328125 | max allocated: 22586.583984375 | reserved: 23360.0 | max reserved: 23360.0 7: iteration 10/ 21553 | consumed samples: 2560 | consumed tokens: 5242880 | elapsed time per iteration (s): 1.34 | learning rate: 9.279E-06 | global batch size: 256 | lm loss: 1.065744E+01 | grad norm: 5.266 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 191.361 | TFLOPs: 6.70 | 7: iteration 20/ 21553 | consumed samples: 5120 | consumed tokens: 10485760 | elapsed time per iteration (s): 0.31 | learning rate: 1.856E-05 | global batch size: 256 | lm loss: 9.736977E+00 | grad norm: 2.160 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 825.830 | TFLOPs: 28.91 | 7: iteration 30/ 21553 | consumed samples: 7680 | consumed tokens: 15728640 | elapsed time per iteration (s): 0.32 | learning rate: 2.784E-05 | global batch size: 256 | lm loss: 9.362369E+00 | grad norm: 2.052 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 807.558 | TFLOPs: 28.27 | 7: iteration 40/ 21553 | consumed samples: 10240 | consumed tokens: 20971520 | elapsed time per iteration (s): 0.31 | learning rate: 3.712E-05 | global batch size: 256 | lm loss: 8.966187E+00 | grad norm: 1.728 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 828.276 | TFLOPs: 29.00 | 7: iteration 50/ 21553 | consumed samples: 12800 | consumed tokens: 26214400 | elapsed time per iteration (s): 0.32 | learning rate: 4.640E-05 | global batch size: 256 | lm loss: 8.554568E+00 | grad norm: 1.545 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 812.618 | TFLOPs: 28.45 | 7: iteration 60/ 21553 | consumed samples: 15360 | consumed tokens: 31457280 | elapsed time per iteration (s): 0.32 | learning rate: 5.568E-05 | global batch size: 256 | lm loss: 8.118588E+00 | grad norm: 1.228 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 809.106 | TFLOPs: 28.32 | 7: iteration 70/ 21553 | consumed samples: 17920 | consumed tokens: 36700160 | elapsed time per iteration (s): 0.32 | learning rate: 6.496E-05 | global batch size: 256 | lm loss: 7.707132E+00 | grad norm: 0.941 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 794.697 | TFLOPs: 27.82 | 7: iteration 80/ 21553 | consumed samples: 20480 | consumed tokens: 41943040 | elapsed time per iteration (s): 0.34 | learning rate: 7.424E-05 | global batch size: 256 | lm loss: 7.348476E+00 | grad norm: 1.035 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 763.271 | TFLOPs: 26.72 | 7: iteration 90/ 21553 | consumed samples: 23040 | consumed tokens: 47185920 | elapsed time per iteration (s): 0.33 | learning rate: 8.351E-05 | global batch size: 256 | lm loss: 7.082109E+00 | grad norm: 0.597 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 779.908 | TFLOPs: 27.30 | 7: iteration 100/ 21553 | consumed samples: 25600 | consumed tokens: 52428800 | elapsed time per iteration (s): 0.33 | learning rate: 9.279E-05 | global batch size: 256 | lm loss: 6.932301E+00 | grad norm: 0.772 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 783.250 | TFLOPs: 27.42 | 7: iteration 110/ 21553 | consumed samples: 28160 | consumed tokens: 57671680 | elapsed time per iteration (s): 0.31 | learning rate: 1.021E-04 | global batch size: 256 | lm loss: 6.790475E+00 | grad norm: 0.942 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 814.111 | TFLOPs: 28.50 | 7: iteration 120/ 21553 | consumed samples: 30720 | consumed tokens: 62914560 | elapsed time per iteration (s): 0.32 | learning rate: 1.114E-04 | global batch size: 256 | lm loss: 6.688376E+00 | grad norm: 0.570 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 809.046 | TFLOPs: 28.32 | 7: iteration 130/ 21553 | consumed samples: 33280 | consumed tokens: 68157440 | elapsed time per iteration (s): 0.31 | learning rate: 1.206E-04 | global batch size: 256 | lm loss: 6.597569E+00 | grad norm: 0.865 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 824.428 | TFLOPs: 28.86 | 7: iteration 140/ 21553 | consumed samples: 35840 | consumed tokens: 73400320 | elapsed time per iteration (s): 0.32 | learning rate: 1.299E-04 | global batch size: 256 | lm loss: 6.508557E+00 | grad norm: 0.444 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 808.832 | TFLOPs: 28.31 | 7: iteration 150/ 21553 | consumed samples: 38400 | consumed tokens: 78643200 | elapsed time per iteration (s): 0.34 | learning rate: 1.392E-04 | global batch size: 256 | lm loss: 6.452471E+00 | grad norm: 0.651 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 745.297 | TFLOPs: 26.09 | 7: iteration 160/ 21553 | consumed samples: 40960 | consumed tokens: 83886080 | elapsed time per iteration (s): 0.30 | learning rate: 1.485E-04 | global batch size: 256 | lm loss: 6.394907E+00 | grad norm: 0.506 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 847.394 | TFLOPs: 29.66 | 7: iteration 170/ 21553 | consumed samples: 43520 | consumed tokens: 89128960 | elapsed time per iteration (s): 0.32 | learning rate: 1.577E-04 | global batch size: 256 | lm loss: 6.344473E+00 | grad norm: 0.814 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 801.242 | TFLOPs: 28.05 | 7: iteration 180/ 21553 | consumed samples: 46080 | consumed tokens: 94371840 | elapsed time per iteration (s): 0.32 | learning rate: 1.670E-04 | global batch size: 256 | lm loss: 6.305540E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 793.943 | TFLOPs: 27.79 | 7: iteration 190/ 21553 | consumed samples: 48640 | consumed tokens: 99614720 | elapsed time per iteration (s): 0.33 | learning rate: 1.763E-04 | global batch size: 256 | lm loss: 6.286691E+00 | grad norm: 0.874 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 775.353 | TFLOPs: 27.14 | 7: iteration 200/ 21553 | consumed samples: 51200 | consumed tokens: 104857600 | elapsed time per iteration (s): 0.32 | learning rate: 1.856E-04 | global batch size: 256 | lm loss: 6.250045E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 789.557 | TFLOPs: 27.64 | 7: iteration 210/ 21553 | consumed samples: 53760 | consumed tokens: 110100480 | elapsed time per iteration (s): 0.33 | learning rate: 1.949E-04 | global batch size: 256 | lm loss: 6.217243E+00 | grad norm: 0.560 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 771.134 | TFLOPs: 27.00 | 7: iteration 220/ 21553 | consumed samples: 56320 | consumed tokens: 115343360 | elapsed time per iteration (s): 0.32 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.170142E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 810.014 | TFLOPs: 28.36 | 7: iteration 230/ 21553 | consumed samples: 58880 | consumed tokens: 120586240 | elapsed time per iteration (s): 0.32 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.152243E+00 | grad norm: 0.541 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 804.616 | TFLOPs: 28.17 | 7: iteration 240/ 21553 | consumed samples: 61440 | consumed tokens: 125829120 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.123818E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 821.673 | TFLOPs: 28.76 | 7: iteration 250/ 21553 | consumed samples: 64000 | consumed tokens: 131072000 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.100248E+00 | grad norm: 0.465 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 845.543 | TFLOPs: 29.60 | 7: iteration 260/ 21553 | consumed samples: 66560 | consumed tokens: 136314880 | elapsed time per iteration (s): 0.32 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.081143E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 808.025 | TFLOPs: 28.29 | 7: iteration 270/ 21553 | consumed samples: 69120 | consumed tokens: 141557760 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.066018E+00 | grad norm: 0.582 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 834.196 | TFLOPs: 29.20 | 7: iteration 280/ 21553 | consumed samples: 71680 | consumed tokens: 146800640 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.056438E+00 | grad norm: 0.639 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 824.406 | TFLOPs: 28.86 | 7: iteration 290/ 21553 | consumed samples: 74240 | consumed tokens: 152043520 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.016298E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 834.068 | TFLOPs: 29.20 | 7: iteration 300/ 21553 | consumed samples: 76800 | consumed tokens: 157286400 | elapsed time per iteration (s): 0.32 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.006890E+00 | grad norm: 0.539 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 791.378 | TFLOPs: 27.70 | 7: iteration 310/ 21553 | consumed samples: 79360 | consumed tokens: 162529280 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 6.021267E+00 | grad norm: 0.675 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 831.447 | TFLOPs: 29.11 | 7: iteration 320/ 21553 | consumed samples: 81920 | consumed tokens: 167772160 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.990482E+00 | grad norm: 0.335 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 820.017 | TFLOPs: 28.71 | 7: iteration 330/ 21553 | consumed samples: 84480 | consumed tokens: 173015040 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.951902E+00 | grad norm: 0.331 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 830.105 | TFLOPs: 29.06 | 7: iteration 340/ 21553 | consumed samples: 87040 | consumed tokens: 178257920 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.921717E+00 | grad norm: 1.132 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 839.522 | TFLOPs: 29.39 | 7: iteration 350/ 21553 | consumed samples: 89600 | consumed tokens: 183500800 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.919493E+00 | grad norm: 0.386 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 816.807 | TFLOPs: 28.59 | 7: iteration 360/ 21553 | consumed samples: 92160 | consumed tokens: 188743680 | elapsed time per iteration (s): 0.32 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.887679E+00 | grad norm: 0.994 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 808.850 | TFLOPs: 28.32 | 7: iteration 370/ 21553 | consumed samples: 94720 | consumed tokens: 193986560 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.868382E+00 | grad norm: 0.663 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 813.053 | TFLOPs: 28.46 | 7: iteration 380/ 21553 | consumed samples: 97280 | consumed tokens: 199229440 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.822472E+00 | grad norm: 0.564 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 814.502 | TFLOPs: 28.51 | 7: iteration 390/ 21553 | consumed samples: 99840 | consumed tokens: 204472320 | elapsed time per iteration (s): 0.32 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.812465E+00 | grad norm: 0.849 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 807.831 | TFLOPs: 28.28 | 7: iteration 400/ 21553 | consumed samples: 102400 | consumed tokens: 209715200 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.788492E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 820.654 | TFLOPs: 28.73 | 7: iteration 410/ 21553 | consumed samples: 104960 | consumed tokens: 214958080 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.775897E+00 | grad norm: 1.033 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 846.849 | TFLOPs: 29.65 | 7: iteration 420/ 21553 | consumed samples: 107520 | consumed tokens: 220200960 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.764762E+00 | grad norm: 0.366 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 816.628 | TFLOPs: 28.59 | 7: iteration 430/ 21553 | consumed samples: 110080 | consumed tokens: 225443840 | elapsed time per iteration (s): 0.32 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.732855E+00 | grad norm: 0.623 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 812.395 | TFLOPs: 28.44 | 7: iteration 440/ 21553 | consumed samples: 112640 | consumed tokens: 230686720 | elapsed time per iteration (s): 0.31 | learning rate: 2.000E-04 | global batch size: 256 | lm loss: 5.707626E+00 | grad norm: 0.540 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 830.930 | TFLOPs: 29.09 | 7: iteration 450/ 21553 | consumed samples: 115200 | consumed tokens: 235929600 | elapsed time per iteration (s): 0.30 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.682688E+00 | grad norm: 0.785 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 844.327 | TFLOPs: 29.56 | 7: iteration 460/ 21553 | consumed samples: 117760 | consumed tokens: 241172480 | elapsed time per iteration (s): 0.31 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.663973E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 815.005 | TFLOPs: 28.53 | 7: iteration 470/ 21553 | consumed samples: 120320 | consumed tokens: 246415360 | elapsed time per iteration (s): 0.31 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.624310E+00 | grad norm: 0.563 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 821.154 | TFLOPs: 28.75 | 7: iteration 480/ 21553 | consumed samples: 122880 | consumed tokens: 251658240 | elapsed time per iteration (s): 0.32 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.625353E+00 | grad norm: 0.579 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 793.635 | TFLOPs: 27.78 | 7: iteration 490/ 21553 | consumed samples: 125440 | consumed tokens: 256901120 | elapsed time per iteration (s): 0.31 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.595184E+00 | grad norm: 1.078 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 821.739 | TFLOPs: 28.77 | 7: iteration 500/ 21553 | consumed samples: 128000 | consumed tokens: 262144000 | elapsed time per iteration (s): 0.32 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.581264E+00 | grad norm: 0.664 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 811.255 | TFLOPs: 28.40 | 7: iteration 510/ 21553 | consumed samples: 130560 | consumed tokens: 267386880 | elapsed time per iteration (s): 0.32 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.566111E+00 | grad norm: 0.747 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 802.738 | TFLOPs: 28.10 | 7: iteration 520/ 21553 | consumed samples: 133120 | consumed tokens: 272629760 | elapsed time per iteration (s): 0.32 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.534502E+00 | grad norm: 0.603 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 801.360 | TFLOPs: 28.05 | 7: iteration 530/ 21553 | consumed samples: 135680 | consumed tokens: 277872640 | elapsed time per iteration (s): 0.30 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.516842E+00 | grad norm: 0.744 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 847.238 | TFLOPs: 29.66 | 7: iteration 540/ 21553 | consumed samples: 138240 | consumed tokens: 283115520 | elapsed time per iteration (s): 0.31 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.491246E+00 | grad norm: 0.942 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 824.596 | TFLOPs: 28.87 | 7: iteration 550/ 21553 | consumed samples: 140800 | consumed tokens: 288358400 | elapsed time per iteration (s): 0.31 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.471751E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 830.785 | TFLOPs: 29.08 | 7: iteration 560/ 21553 | consumed samples: 143360 | consumed tokens: 293601280 | elapsed time per iteration (s): 0.32 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.468285E+00 | grad norm: 0.719 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 810.698 | TFLOPs: 28.38 | 7: iteration 570/ 21553 | consumed samples: 145920 | consumed tokens: 298844160 | elapsed time per iteration (s): 0.30 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.432404E+00 | grad norm: 0.654 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 845.011 | TFLOPs: 29.58 | 7: iteration 580/ 21553 | consumed samples: 148480 | consumed tokens: 304087040 | elapsed time per iteration (s): 0.31 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.410930E+00 | grad norm: 0.575 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 826.772 | TFLOPs: 28.94 | 7: iteration 590/ 21553 | consumed samples: 151040 | consumed tokens: 309329920 | elapsed time per iteration (s): 0.31 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.407993E+00 | grad norm: 0.734 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 819.127 | TFLOPs: 28.68 | 7: iteration 600/ 21553 | consumed samples: 153600 | consumed tokens: 314572800 | elapsed time per iteration (s): 0.30 | learning rate: 1.999E-04 | global batch size: 256 | lm loss: 5.393534E+00 | grad norm: 0.609 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 851.136 | TFLOPs: 29.80 | 7: iteration 610/ 21553 | consumed samples: 156160 | consumed tokens: 319815680 | elapsed time per iteration (s): 0.31 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.354587E+00 | grad norm: 0.471 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 834.257 | TFLOPs: 29.21 | 7: iteration 620/ 21553 | consumed samples: 158720 | consumed tokens: 325058560 | elapsed time per iteration (s): 0.30 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.341993E+00 | grad norm: 0.836 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 846.106 | TFLOPs: 29.62 | 7: iteration 630/ 21553 | consumed samples: 161280 | consumed tokens: 330301440 | elapsed time per iteration (s): 0.30 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.320016E+00 | grad norm: 0.700 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 843.448 | TFLOPs: 29.53 | 7: iteration 640/ 21553 | consumed samples: 163840 | consumed tokens: 335544320 | elapsed time per iteration (s): 0.31 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.298060E+00 | grad norm: 0.606 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 819.566 | TFLOPs: 28.69 | 7: iteration 650/ 21553 | consumed samples: 166400 | consumed tokens: 340787200 | elapsed time per iteration (s): 0.31 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.276625E+00 | grad norm: 0.739 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 827.372 | TFLOPs: 28.96 | 7: iteration 660/ 21553 | consumed samples: 168960 | consumed tokens: 346030080 | elapsed time per iteration (s): 0.30 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.268057E+00 | grad norm: 0.888 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 844.264 | TFLOPs: 29.56 | 7: iteration 670/ 21553 | consumed samples: 171520 | consumed tokens: 351272960 | elapsed time per iteration (s): 0.31 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.255247E+00 | grad norm: 0.685 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 835.928 | TFLOPs: 29.26 | 7: iteration 680/ 21553 | consumed samples: 174080 | consumed tokens: 356515840 | elapsed time per iteration (s): 0.31 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.228270E+00 | grad norm: 0.645 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 818.325 | TFLOPs: 28.65 | 7: iteration 690/ 21553 | consumed samples: 176640 | consumed tokens: 361758720 | elapsed time per iteration (s): 0.32 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.213612E+00 | grad norm: 0.732 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 798.872 | TFLOPs: 27.97 | 7: iteration 700/ 21553 | consumed samples: 179200 | consumed tokens: 367001600 | elapsed time per iteration (s): 0.30 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.198479E+00 | grad norm: 0.689 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 847.536 | TFLOPs: 29.67 | 7: iteration 710/ 21553 | consumed samples: 181760 | consumed tokens: 372244480 | elapsed time per iteration (s): 0.30 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.185516E+00 | grad norm: 0.710 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 850.678 | TFLOPs: 29.78 | 7: iteration 720/ 21553 | consumed samples: 184320 | consumed tokens: 377487360 | elapsed time per iteration (s): 0.30 | learning rate: 1.998E-04 | global batch size: 256 | lm loss: 5.176759E+00 | grad norm: 0.633 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 842.959 | TFLOPs: 29.51 | 7: iteration 730/ 21553 | consumed samples: 186880 | consumed tokens: 382730240 | elapsed time per iteration (s): 0.31 | learning rate: 1.997E-04 | global batch size: 256 | lm loss: 5.154255E+00 | grad norm: 0.561 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 833.862 | TFLOPs: 29.19 | 7: iteration 740/ 21553 | consumed samples: 189440 | consumed tokens: 387973120 | elapsed time per iteration (s): 0.32 | learning rate: 1.997E-04 | global batch size: 256 | lm loss: 5.140783E+00 | grad norm: 0.626 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 806.029 | TFLOPs: 28.22 | 7: iteration 750/ 21553 | consumed samples: 192000 | consumed tokens: 393216000 | elapsed time per iteration (s): 0.31 | learning rate: 1.997E-04 | global batch size: 256 | lm loss: 5.127533E+00 | grad norm: 0.780 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 824.942 | TFLOPs: 28.88 | 7: iteration 760/ 21553 | consumed samples: 194560 | consumed tokens: 398458880 | elapsed time per iteration (s): 0.30 | learning rate: 1.997E-04 | global batch size: 256 | lm loss: 5.120929E+00 | grad norm: 0.607 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 851.423 | TFLOPs: 29.81 | 7: iteration 770/ 21553 | consumed samples: 197120 | consumed tokens: 403701760 | elapsed time per iteration (s): 0.30 | learning rate: 1.997E-04 | global batch size: 256 | lm loss: 5.091842E+00 | grad norm: 0.856 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.558 | TFLOPs: 30.02 | 7: iteration 780/ 21553 | consumed samples: 199680 | consumed tokens: 408944640 | elapsed time per iteration (s): 0.31 | learning rate: 1.997E-04 | global batch size: 256 | lm loss: 5.079604E+00 | grad norm: 0.652 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 836.516 | TFLOPs: 29.28 | 7: iteration 790/ 21553 | consumed samples: 202240 | consumed tokens: 414187520 | elapsed time per iteration (s): 0.32 | learning rate: 1.997E-04 | global batch size: 256 | lm loss: 5.069685E+00 | grad norm: 0.548 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 796.142 | TFLOPs: 27.87 | 7: iteration 800/ 21553 | consumed samples: 204800 | consumed tokens: 419430400 | elapsed time per iteration (s): 0.30 | learning rate: 1.997E-04 | global batch size: 256 | lm loss: 5.046468E+00 | grad norm: 0.621 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 847.349 | TFLOPs: 29.66 | 7: iteration 810/ 21553 | consumed samples: 207360 | consumed tokens: 424673280 | elapsed time per iteration (s): 0.30 | learning rate: 1.997E-04 | global batch size: 256 | lm loss: 5.037503E+00 | grad norm: 0.627 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 840.090 | TFLOPs: 29.41 | 7: iteration 820/ 21553 | consumed samples: 209920 | consumed tokens: 429916160 | elapsed time per iteration (s): 0.32 | learning rate: 1.996E-04 | global batch size: 256 | lm loss: 5.038281E+00 | grad norm: 0.909 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 788.677 | TFLOPs: 27.61 | 7: iteration 830/ 21553 | consumed samples: 212480 | consumed tokens: 435159040 | elapsed time per iteration (s): 0.32 | learning rate: 1.996E-04 | global batch size: 256 | lm loss: 5.017791E+00 | grad norm: 0.615 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 790.724 | TFLOPs: 27.68 | 7: iteration 840/ 21553 | consumed samples: 215040 | consumed tokens: 440401920 | elapsed time per iteration (s): 0.30 | learning rate: 1.996E-04 | global batch size: 256 | lm loss: 5.005743E+00 | grad norm: 0.943 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 849.268 | TFLOPs: 29.73 | 7: iteration 850/ 21553 | consumed samples: 217600 | consumed tokens: 445644800 | elapsed time per iteration (s): 0.30 | learning rate: 1.996E-04 | global batch size: 256 | lm loss: 5.007981E+00 | grad norm: 0.644 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.862 | TFLOPs: 30.07 | 7: iteration 860/ 21553 | consumed samples: 220160 | consumed tokens: 450887680 | elapsed time per iteration (s): 0.29 | learning rate: 1.996E-04 | global batch size: 256 | lm loss: 4.983039E+00 | grad norm: 0.481 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 869.178 | TFLOPs: 30.43 | 7: iteration 870/ 21553 | consumed samples: 222720 | consumed tokens: 456130560 | elapsed time per iteration (s): 0.30 | learning rate: 1.996E-04 | global batch size: 256 | lm loss: 4.968596E+00 | grad norm: 0.868 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 841.440 | TFLOPs: 29.46 | 7: iteration 880/ 21553 | consumed samples: 225280 | consumed tokens: 461373440 | elapsed time per iteration (s): 0.30 | learning rate: 1.996E-04 | global batch size: 256 | lm loss: 4.963697E+00 | grad norm: 0.660 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.489 | TFLOPs: 30.23 | 7: iteration 890/ 21553 | consumed samples: 227840 | consumed tokens: 466616320 | elapsed time per iteration (s): 0.31 | learning rate: 1.996E-04 | global batch size: 256 | lm loss: 4.953908E+00 | grad norm: 0.808 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 830.503 | TFLOPs: 29.07 | 7: iteration 900/ 21553 | consumed samples: 230400 | consumed tokens: 471859200 | elapsed time per iteration (s): 0.33 | learning rate: 1.995E-04 | global batch size: 256 | lm loss: 4.926603E+00 | grad norm: 0.563 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 787.114 | TFLOPs: 27.55 | 7: iteration 910/ 21553 | consumed samples: 232960 | consumed tokens: 477102080 | elapsed time per iteration (s): 0.30 | learning rate: 1.995E-04 | global batch size: 256 | lm loss: 4.924453E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.849 | TFLOPs: 30.24 | 7: iteration 920/ 21553 | consumed samples: 235520 | consumed tokens: 482344960 | elapsed time per iteration (s): 0.30 | learning rate: 1.995E-04 | global batch size: 256 | lm loss: 4.905740E+00 | grad norm: 0.753 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 844.078 | TFLOPs: 29.55 | 7: iteration 930/ 21553 | consumed samples: 238080 | consumed tokens: 487587840 | elapsed time per iteration (s): 0.30 | learning rate: 1.995E-04 | global batch size: 256 | lm loss: 4.908614E+00 | grad norm: 0.701 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 848.493 | TFLOPs: 29.70 | 7: iteration 940/ 21553 | consumed samples: 240640 | consumed tokens: 492830720 | elapsed time per iteration (s): 0.29 | learning rate: 1.995E-04 | global batch size: 256 | lm loss: 4.888618E+00 | grad norm: 1.019 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.804 | TFLOPs: 30.38 | 7: iteration 950/ 21553 | consumed samples: 243200 | consumed tokens: 498073600 | elapsed time per iteration (s): 0.31 | learning rate: 1.995E-04 | global batch size: 256 | lm loss: 4.879775E+00 | grad norm: 0.658 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 836.987 | TFLOPs: 29.30 | 7: iteration 960/ 21553 | consumed samples: 245760 | consumed tokens: 503316480 | elapsed time per iteration (s): 0.31 | learning rate: 1.995E-04 | global batch size: 256 | lm loss: 4.860705E+00 | grad norm: 0.670 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 834.503 | TFLOPs: 29.21 | 7: iteration 970/ 21553 | consumed samples: 248320 | consumed tokens: 508559360 | elapsed time per iteration (s): 0.30 | learning rate: 1.994E-04 | global batch size: 256 | lm loss: 4.845752E+00 | grad norm: 0.700 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 860.310 | TFLOPs: 30.12 | 7: iteration 980/ 21553 | consumed samples: 250880 | consumed tokens: 513802240 | elapsed time per iteration (s): 0.30 | learning rate: 1.994E-04 | global batch size: 256 | lm loss: 4.853506E+00 | grad norm: 0.545 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.103 | TFLOPs: 29.97 | 7: iteration 990/ 21553 | consumed samples: 253440 | consumed tokens: 519045120 | elapsed time per iteration (s): 0.29 | learning rate: 1.994E-04 | global batch size: 256 | lm loss: 4.824920E+00 | grad norm: 0.660 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.640 | TFLOPs: 30.41 | 7: iteration 1000/ 21553 | consumed samples: 256000 | consumed tokens: 524288000 | elapsed time per iteration (s): 0.30 | learning rate: 1.994E-04 | global batch size: 256 | lm loss: 4.822888E+00 | grad norm: 0.682 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 851.063 | TFLOPs: 29.79 | 7: ----------------------------------------------------------------------------------------------- 7: validation loss at iteration 1000 | lm loss value: 4.813007E+00 | lm loss PPL: 1.231012E+02 | 7: ----------------------------------------------------------------------------------------------- 0: saving checkpoint at iteration 1000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:00:58,940] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step1000 is begin to save! 0: [2023-03-17 00:00:59,044] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:00:59,133] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:00:59,133] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:00:59,150] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:00:59,150] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:00:59,165] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:00:59,165] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:00:59,180] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:00:59,180] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:00:59,195] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:00:59,195] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:00:59,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:00:59,210] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:00:59,224] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:00:59,224] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:00:59,239] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:00:59,239] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:00:59,254] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:00:59,254] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:00:59,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:00:59,269] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:00:59,283] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:00:59,283] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:00:59,298] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:00:59,298] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:00:59,313] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:00:59,313] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:00:59,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:00:59,328] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:00:59,342] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:00:59,343] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:00:59,357] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:00:59,357] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:00:59,358] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:00:59,359] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step1000/mp_rank_00_model_states.pt 0: [2023-03-17 00:00:59,359] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:00:59,362] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:00:59,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:00:59,414] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:00:59,416] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:00:59,416] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 00:00:59,416] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 0: [2023-03-17 00:00:59,416] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:00:59,416] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 00:00:59,416] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 0: [2023-03-17 00:00:59,417] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:00:59,417] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:00:59,417] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 0: [2023-03-17 00:00:59,417] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:00:59,418] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 00:00:59,418] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 0: [2023-03-17 00:00:59,419] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:00:59,419] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 00:00:59,419] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 0: [2023-03-17 00:00:59,420] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:00:59,421] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:00:59,421] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 00:00:59,421] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 00:00:59,421] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 0: [2023-03-17 00:00:59,421] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 6: [2023-03-17 00:00:59,425] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:00:59,425] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,425] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 2: [2023-03-17 00:00:59,430] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,430] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,430] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,430] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:00:59,430] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:00:59,430] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 2: [2023-03-17 00:00:59,430] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 2: [2023-03-17 00:00:59,430] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:00:59,430] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 7: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 7: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 7: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 7: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 2: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 2: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 1: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:00:59,434] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 00:00:59,434] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:00:59,434] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:00:59,434] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 1: [2023-03-17 00:00:59,434] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 1: [2023-03-17 00:00:59,434] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 3: [2023-03-17 00:00:59,434] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:00:59,435] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:00:59,435] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 3: [2023-03-17 00:00:59,435] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:00:59,435] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 00:00:59,436] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 3: [2023-03-17 00:00:59,436] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:00:59,436] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 00:00:59,436] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 3: [2023-03-17 00:00:59,436] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:00:59,436] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 00:00:59,436] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 4: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 00:00:59,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 3: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 3: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:00:59,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:00:59,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:00:59,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 3: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 4: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 3: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 3: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 3: [2023-03-17 00:00:59,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 00:00:59,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 5: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:00:59,438] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:00:59,438] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:00:59,438] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 5: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 5: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 6: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:00:59,433] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 6: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:00:59,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 00:00:59,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:00:59,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 00:00:59,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:00:59,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 6: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 5: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 6: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 6: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 6: [2023-03-17 00:00:59,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:00:59,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 4: [2023-03-17 00:00:59,440] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,440] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,440] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,440] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,440] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:00:59,440] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 00:00:59,440] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 00:00:59,440] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:00:59,440] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 4: [2023-03-17 00:00:59,440] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 4: [2023-03-17 00:00:59,440] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 4: [2023-03-17 00:00:59,440] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 4: [2023-03-17 00:00:59,441] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:00:59,441] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 00:00:59,441] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 0: [2023-03-17 00:00:59,445] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step1000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:00:59,445] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step1000 is ready now! 0: successfully saved checkpoint at iteration 1000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 512.29 7: iteration 1010/ 21553 | consumed samples: 258560 | consumed tokens: 529530880 | elapsed time per iteration (s): 0.36 | learning rate: 1.994E-04 | global batch size: 256 | lm loss: 4.802536E+00 | grad norm: 0.941 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 717.785 | TFLOPs: 25.13 | 7: iteration 1020/ 21553 | consumed samples: 261120 | consumed tokens: 534773760 | elapsed time per iteration (s): 0.30 | learning rate: 1.994E-04 | global batch size: 256 | lm loss: 4.798707E+00 | grad norm: 0.842 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.335 | TFLOPs: 30.33 | 7: iteration 1030/ 21553 | consumed samples: 263680 | consumed tokens: 540016640 | elapsed time per iteration (s): 0.30 | learning rate: 1.994E-04 | global batch size: 256 | lm loss: 4.809159E+00 | grad norm: 0.539 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.704 | TFLOPs: 30.34 | 7: iteration 1040/ 21553 | consumed samples: 266240 | consumed tokens: 545259520 | elapsed time per iteration (s): 0.30 | learning rate: 1.993E-04 | global batch size: 256 | lm loss: 4.778893E+00 | grad norm: 0.869 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.340 | TFLOPs: 30.36 | 7: iteration 1050/ 21553 | consumed samples: 268800 | consumed tokens: 550502400 | elapsed time per iteration (s): 0.30 | learning rate: 1.993E-04 | global batch size: 256 | lm loss: 4.777409E+00 | grad norm: 0.504 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 851.137 | TFLOPs: 29.80 | 7: iteration 1060/ 21553 | consumed samples: 271360 | consumed tokens: 555745280 | elapsed time per iteration (s): 0.30 | learning rate: 1.993E-04 | global batch size: 256 | lm loss: 4.749527E+00 | grad norm: 0.791 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.272 | TFLOPs: 30.36 | 7: iteration 1070/ 21553 | consumed samples: 273920 | consumed tokens: 560988160 | elapsed time per iteration (s): 0.30 | learning rate: 1.993E-04 | global batch size: 256 | lm loss: 4.735986E+00 | grad norm: 0.840 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.519 | TFLOPs: 30.37 | 7: iteration 1080/ 21553 | consumed samples: 276480 | consumed tokens: 566231040 | elapsed time per iteration (s): 0.29 | learning rate: 1.993E-04 | global batch size: 256 | lm loss: 4.746076E+00 | grad norm: 0.609 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.000 | TFLOPs: 30.39 | 7: iteration 1090/ 21553 | consumed samples: 279040 | consumed tokens: 571473920 | elapsed time per iteration (s): 0.30 | learning rate: 1.993E-04 | global batch size: 256 | lm loss: 4.720252E+00 | grad norm: 0.615 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.638 | TFLOPs: 30.37 | 7: iteration 1100/ 21553 | consumed samples: 281600 | consumed tokens: 576716800 | elapsed time per iteration (s): 0.30 | learning rate: 1.992E-04 | global batch size: 256 | lm loss: 4.718612E+00 | grad norm: 0.692 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.279 | TFLOPs: 30.36 | 7: iteration 1110/ 21553 | consumed samples: 284160 | consumed tokens: 581959680 | elapsed time per iteration (s): 0.30 | learning rate: 1.992E-04 | global batch size: 256 | lm loss: 4.704027E+00 | grad norm: 0.605 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 845.222 | TFLOPs: 29.59 | 7: iteration 1120/ 21553 | consumed samples: 286720 | consumed tokens: 587202560 | elapsed time per iteration (s): 0.30 | learning rate: 1.992E-04 | global batch size: 256 | lm loss: 4.688346E+00 | grad norm: 0.654 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.920 | TFLOPs: 30.21 | 7: iteration 1130/ 21553 | consumed samples: 289280 | consumed tokens: 592445440 | elapsed time per iteration (s): 0.30 | learning rate: 1.992E-04 | global batch size: 256 | lm loss: 4.683709E+00 | grad norm: 0.784 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.630 | TFLOPs: 30.37 | 7: iteration 1140/ 21553 | consumed samples: 291840 | consumed tokens: 597688320 | elapsed time per iteration (s): 0.30 | learning rate: 1.992E-04 | global batch size: 256 | lm loss: 4.671601E+00 | grad norm: 0.591 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.494 | TFLOPs: 30.37 | 7: iteration 1150/ 21553 | consumed samples: 294400 | consumed tokens: 602931200 | elapsed time per iteration (s): 0.30 | learning rate: 1.991E-04 | global batch size: 256 | lm loss: 4.666644E+00 | grad norm: 0.713 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.340 | TFLOPs: 30.36 | 7: iteration 1160/ 21553 | consumed samples: 296960 | consumed tokens: 608174080 | elapsed time per iteration (s): 0.30 | learning rate: 1.991E-04 | global batch size: 256 | lm loss: 4.660798E+00 | grad norm: 0.589 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.077 | TFLOPs: 30.35 | 7: iteration 1170/ 21553 | consumed samples: 299520 | consumed tokens: 613416960 | elapsed time per iteration (s): 0.30 | learning rate: 1.991E-04 | global batch size: 256 | lm loss: 4.657310E+00 | grad norm: 0.740 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.911 | TFLOPs: 30.35 | 7: iteration 1180/ 21553 | consumed samples: 302080 | consumed tokens: 618659840 | elapsed time per iteration (s): 0.30 | learning rate: 1.991E-04 | global batch size: 256 | lm loss: 4.653234E+00 | grad norm: 0.589 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.422 | TFLOPs: 30.37 | 7: iteration 1190/ 21553 | consumed samples: 304640 | consumed tokens: 623902720 | elapsed time per iteration (s): 0.30 | learning rate: 1.991E-04 | global batch size: 256 | lm loss: 4.631143E+00 | grad norm: 0.644 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.616 | TFLOPs: 29.99 | 7: iteration 1200/ 21553 | consumed samples: 307200 | consumed tokens: 629145600 | elapsed time per iteration (s): 0.30 | learning rate: 1.991E-04 | global batch size: 256 | lm loss: 4.619937E+00 | grad norm: 0.657 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.948 | TFLOPs: 30.35 | 7: iteration 1210/ 21553 | consumed samples: 309760 | consumed tokens: 634388480 | elapsed time per iteration (s): 0.30 | learning rate: 1.990E-04 | global batch size: 256 | lm loss: 4.616426E+00 | grad norm: 0.552 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.055 | TFLOPs: 30.18 | 7: iteration 1220/ 21553 | consumed samples: 312320 | consumed tokens: 639631360 | elapsed time per iteration (s): 0.30 | learning rate: 1.990E-04 | global batch size: 256 | lm loss: 4.619030E+00 | grad norm: 0.772 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.282 | TFLOPs: 30.36 | 7: iteration 1230/ 21553 | consumed samples: 314880 | consumed tokens: 644874240 | elapsed time per iteration (s): 0.30 | learning rate: 1.990E-04 | global batch size: 256 | lm loss: 4.608610E+00 | grad norm: 0.744 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.561 | TFLOPs: 30.37 | 7: iteration 1240/ 21553 | consumed samples: 317440 | consumed tokens: 650117120 | elapsed time per iteration (s): 0.30 | learning rate: 1.990E-04 | global batch size: 256 | lm loss: 4.599579E+00 | grad norm: 0.675 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.747 | TFLOPs: 30.38 | 7: iteration 1250/ 21553 | consumed samples: 320000 | consumed tokens: 655360000 | elapsed time per iteration (s): 0.30 | learning rate: 1.990E-04 | global batch size: 256 | lm loss: 4.593883E+00 | grad norm: 0.518 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.210 | TFLOPs: 30.36 | 7: iteration 1260/ 21553 | consumed samples: 322560 | consumed tokens: 660602880 | elapsed time per iteration (s): 0.30 | learning rate: 1.989E-04 | global batch size: 256 | lm loss: 4.582605E+00 | grad norm: 0.521 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.379 | TFLOPs: 30.36 | 7: iteration 1270/ 21553 | consumed samples: 325120 | consumed tokens: 665845760 | elapsed time per iteration (s): 0.30 | learning rate: 1.989E-04 | global batch size: 256 | lm loss: 4.563007E+00 | grad norm: 0.567 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.585 | TFLOPs: 30.37 | 7: iteration 1280/ 21553 | consumed samples: 327680 | consumed tokens: 671088640 | elapsed time per iteration (s): 0.30 | learning rate: 1.989E-04 | global batch size: 256 | lm loss: 4.569727E+00 | grad norm: 0.899 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.328 | TFLOPs: 30.36 | 7: iteration 1290/ 21553 | consumed samples: 330240 | consumed tokens: 676331520 | elapsed time per iteration (s): 0.30 | learning rate: 1.989E-04 | global batch size: 256 | lm loss: 4.568237E+00 | grad norm: 0.713 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.772 | TFLOPs: 30.34 | 7: iteration 1300/ 21553 | consumed samples: 332800 | consumed tokens: 681574400 | elapsed time per iteration (s): 0.30 | learning rate: 1.989E-04 | global batch size: 256 | lm loss: 4.554508E+00 | grad norm: 0.717 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.519 | TFLOPs: 30.37 | 7: iteration 1310/ 21553 | consumed samples: 335360 | consumed tokens: 686817280 | elapsed time per iteration (s): 0.30 | learning rate: 1.988E-04 | global batch size: 256 | lm loss: 4.554206E+00 | grad norm: 0.672 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.977 | TFLOPs: 30.00 | 7: iteration 1320/ 21553 | consumed samples: 337920 | consumed tokens: 692060160 | elapsed time per iteration (s): 0.30 | learning rate: 1.988E-04 | global batch size: 256 | lm loss: 4.547130E+00 | grad norm: 0.510 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.703 | TFLOPs: 30.38 | 7: iteration 1330/ 21553 | consumed samples: 340480 | consumed tokens: 697303040 | elapsed time per iteration (s): 0.30 | learning rate: 1.988E-04 | global batch size: 256 | lm loss: 4.541963E+00 | grad norm: 0.637 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.412 | TFLOPs: 30.37 | 7: iteration 1340/ 21553 | consumed samples: 343040 | consumed tokens: 702545920 | elapsed time per iteration (s): 0.30 | learning rate: 1.988E-04 | global batch size: 256 | lm loss: 4.532430E+00 | grad norm: 0.591 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.053 | TFLOPs: 30.21 | 7: iteration 1350/ 21553 | consumed samples: 345600 | consumed tokens: 707788800 | elapsed time per iteration (s): 0.30 | learning rate: 1.987E-04 | global batch size: 256 | lm loss: 4.517699E+00 | grad norm: 0.554 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.941 | TFLOPs: 30.00 | 7: iteration 1360/ 21553 | consumed samples: 348160 | consumed tokens: 713031680 | elapsed time per iteration (s): 0.30 | learning rate: 1.987E-04 | global batch size: 256 | lm loss: 4.517816E+00 | grad norm: 0.676 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 850.151 | TFLOPs: 29.76 | 7: iteration 1370/ 21553 | consumed samples: 350720 | consumed tokens: 718274560 | elapsed time per iteration (s): 0.30 | learning rate: 1.987E-04 | global batch size: 256 | lm loss: 4.518183E+00 | grad norm: 0.619 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 851.826 | TFLOPs: 29.82 | 7: iteration 1380/ 21553 | consumed samples: 353280 | consumed tokens: 723517440 | elapsed time per iteration (s): 0.30 | learning rate: 1.987E-04 | global batch size: 256 | lm loss: 4.504878E+00 | grad norm: 0.485 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.084 | TFLOPs: 30.35 | 7: iteration 1390/ 21553 | consumed samples: 355840 | consumed tokens: 728760320 | elapsed time per iteration (s): 0.30 | learning rate: 1.987E-04 | global batch size: 256 | lm loss: 4.494106E+00 | grad norm: 0.534 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.350 | TFLOPs: 30.36 | 7: iteration 1400/ 21553 | consumed samples: 358400 | consumed tokens: 734003200 | elapsed time per iteration (s): 0.30 | learning rate: 1.986E-04 | global batch size: 256 | lm loss: 4.485987E+00 | grad norm: 0.489 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.995 | TFLOPs: 30.35 | 7: iteration 1410/ 21553 | consumed samples: 360960 | consumed tokens: 739246080 | elapsed time per iteration (s): 0.30 | learning rate: 1.986E-04 | global batch size: 256 | lm loss: 4.486893E+00 | grad norm: 0.641 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 851.395 | TFLOPs: 29.80 | 7: iteration 1420/ 21553 | consumed samples: 363520 | consumed tokens: 744488960 | elapsed time per iteration (s): 0.30 | learning rate: 1.986E-04 | global batch size: 256 | lm loss: 4.487835E+00 | grad norm: 0.514 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.172 | TFLOPs: 30.36 | 7: iteration 1430/ 21553 | consumed samples: 366080 | consumed tokens: 749731840 | elapsed time per iteration (s): 0.30 | learning rate: 1.986E-04 | global batch size: 256 | lm loss: 4.479491E+00 | grad norm: 0.668 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.795 | TFLOPs: 30.34 | 7: iteration 1440/ 21553 | consumed samples: 368640 | consumed tokens: 754974720 | elapsed time per iteration (s): 0.30 | learning rate: 1.985E-04 | global batch size: 256 | lm loss: 4.480638E+00 | grad norm: 0.715 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.306 | TFLOPs: 30.36 | 7: iteration 1450/ 21553 | consumed samples: 371200 | consumed tokens: 760217600 | elapsed time per iteration (s): 0.30 | learning rate: 1.985E-04 | global batch size: 256 | lm loss: 4.463094E+00 | grad norm: 0.533 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.124 | TFLOPs: 30.36 | 7: iteration 1460/ 21553 | consumed samples: 373760 | consumed tokens: 765460480 | elapsed time per iteration (s): 0.30 | learning rate: 1.985E-04 | global batch size: 256 | lm loss: 4.474129E+00 | grad norm: 0.769 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 859.459 | TFLOPs: 30.09 | 7: iteration 1470/ 21553 | consumed samples: 376320 | consumed tokens: 770703360 | elapsed time per iteration (s): 0.30 | learning rate: 1.985E-04 | global batch size: 256 | lm loss: 4.465340E+00 | grad norm: 0.617 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.747 | TFLOPs: 29.99 | 7: iteration 1480/ 21553 | consumed samples: 378880 | consumed tokens: 775946240 | elapsed time per iteration (s): 0.30 | learning rate: 1.984E-04 | global batch size: 256 | lm loss: 4.459168E+00 | grad norm: 0.694 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.163 | TFLOPs: 30.36 | 7: iteration 1490/ 21553 | consumed samples: 381440 | consumed tokens: 781189120 | elapsed time per iteration (s): 0.30 | learning rate: 1.984E-04 | global batch size: 256 | lm loss: 4.448241E+00 | grad norm: 0.501 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 859.226 | TFLOPs: 30.08 | 7: iteration 1500/ 21553 | consumed samples: 384000 | consumed tokens: 786432000 | elapsed time per iteration (s): 0.30 | learning rate: 1.984E-04 | global batch size: 256 | lm loss: 4.434554E+00 | grad norm: 0.796 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 850.144 | TFLOPs: 29.76 | 7: iteration 1510/ 21553 | consumed samples: 386560 | consumed tokens: 791674880 | elapsed time per iteration (s): 0.30 | learning rate: 1.984E-04 | global batch size: 256 | lm loss: 4.447405E+00 | grad norm: 0.508 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.897 | TFLOPs: 30.35 | 7: iteration 1520/ 21553 | consumed samples: 389120 | consumed tokens: 796917760 | elapsed time per iteration (s): 0.30 | learning rate: 1.983E-04 | global batch size: 256 | lm loss: 4.423865E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.429 | TFLOPs: 30.37 | 7: iteration 1530/ 21553 | consumed samples: 391680 | consumed tokens: 802160640 | elapsed time per iteration (s): 0.30 | learning rate: 1.983E-04 | global batch size: 256 | lm loss: 4.423627E+00 | grad norm: 0.502 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.634 | TFLOPs: 30.37 | 7: iteration 1540/ 21553 | consumed samples: 394240 | consumed tokens: 807403520 | elapsed time per iteration (s): 0.30 | learning rate: 1.983E-04 | global batch size: 256 | lm loss: 4.425559E+00 | grad norm: 0.534 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.766 | TFLOPs: 30.38 | 7: iteration 1550/ 21553 | consumed samples: 396800 | consumed tokens: 812646400 | elapsed time per iteration (s): 0.29 | learning rate: 1.983E-04 | global batch size: 256 | lm loss: 4.414915E+00 | grad norm: 0.579 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.040 | TFLOPs: 30.39 | 7: iteration 1560/ 21553 | consumed samples: 399360 | consumed tokens: 817889280 | elapsed time per iteration (s): 0.30 | learning rate: 1.982E-04 | global batch size: 256 | lm loss: 4.413850E+00 | grad norm: 0.565 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.548 | TFLOPs: 30.37 | 7: iteration 1570/ 21553 | consumed samples: 401920 | consumed tokens: 823132160 | elapsed time per iteration (s): 0.30 | learning rate: 1.982E-04 | global batch size: 256 | lm loss: 4.405863E+00 | grad norm: 0.669 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.299 | TFLOPs: 30.01 | 7: iteration 1580/ 21553 | consumed samples: 404480 | consumed tokens: 828375040 | elapsed time per iteration (s): 0.29 | learning rate: 1.982E-04 | global batch size: 256 | lm loss: 4.403168E+00 | grad norm: 0.567 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.819 | TFLOPs: 30.38 | 7: iteration 1590/ 21553 | consumed samples: 407040 | consumed tokens: 833617920 | elapsed time per iteration (s): 0.30 | learning rate: 1.982E-04 | global batch size: 256 | lm loss: 4.407585E+00 | grad norm: 0.703 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.560 | TFLOPs: 30.37 | 7: iteration 1600/ 21553 | consumed samples: 409600 | consumed tokens: 838860800 | elapsed time per iteration (s): 0.30 | learning rate: 1.981E-04 | global batch size: 256 | lm loss: 4.397497E+00 | grad norm: 0.472 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.124 | TFLOPs: 30.36 | 7: iteration 1610/ 21553 | consumed samples: 412160 | consumed tokens: 844103680 | elapsed time per iteration (s): 0.30 | learning rate: 1.981E-04 | global batch size: 256 | lm loss: 4.396810E+00 | grad norm: 0.531 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.771 | TFLOPs: 30.38 | 7: iteration 1620/ 21553 | consumed samples: 414720 | consumed tokens: 849346560 | elapsed time per iteration (s): 0.30 | learning rate: 1.981E-04 | global batch size: 256 | lm loss: 4.386878E+00 | grad norm: 0.484 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.288 | TFLOPs: 30.36 | 7: iteration 1630/ 21553 | consumed samples: 417280 | consumed tokens: 854589440 | elapsed time per iteration (s): 0.30 | learning rate: 1.981E-04 | global batch size: 256 | lm loss: 4.387323E+00 | grad norm: 0.460 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.729 | TFLOPs: 30.38 | 7: iteration 1640/ 21553 | consumed samples: 419840 | consumed tokens: 859832320 | elapsed time per iteration (s): 0.30 | learning rate: 1.980E-04 | global batch size: 256 | lm loss: 4.370950E+00 | grad norm: 0.559 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.193 | TFLOPs: 30.36 | 7: iteration 1650/ 21553 | consumed samples: 422400 | consumed tokens: 865075200 | elapsed time per iteration (s): 0.30 | learning rate: 1.980E-04 | global batch size: 256 | lm loss: 4.385400E+00 | grad norm: 0.645 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.169 | TFLOPs: 30.36 | 7: iteration 1660/ 21553 | consumed samples: 424960 | consumed tokens: 870318080 | elapsed time per iteration (s): 0.30 | learning rate: 1.980E-04 | global batch size: 256 | lm loss: 4.374905E+00 | grad norm: 0.565 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.263 | TFLOPs: 30.36 | 7: iteration 1670/ 21553 | consumed samples: 427520 | consumed tokens: 875560960 | elapsed time per iteration (s): 0.30 | learning rate: 1.979E-04 | global batch size: 256 | lm loss: 4.368830E+00 | grad norm: 0.542 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.754 | TFLOPs: 30.34 | 7: iteration 1680/ 21553 | consumed samples: 430080 | consumed tokens: 880803840 | elapsed time per iteration (s): 0.30 | learning rate: 1.979E-04 | global batch size: 256 | lm loss: 4.367192E+00 | grad norm: 0.590 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.230 | TFLOPs: 30.36 | 7: iteration 1690/ 21553 | consumed samples: 432640 | consumed tokens: 886046720 | elapsed time per iteration (s): 0.30 | learning rate: 1.979E-04 | global batch size: 256 | lm loss: 4.361133E+00 | grad norm: 0.539 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.364 | TFLOPs: 30.26 | 7: iteration 1700/ 21553 | consumed samples: 435200 | consumed tokens: 891289600 | elapsed time per iteration (s): 0.30 | learning rate: 1.979E-04 | global batch size: 256 | lm loss: 4.347659E+00 | grad norm: 0.536 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.120 | TFLOPs: 30.36 | 7: iteration 1710/ 21553 | consumed samples: 437760 | consumed tokens: 896532480 | elapsed time per iteration (s): 0.29 | learning rate: 1.978E-04 | global batch size: 256 | lm loss: 4.365695E+00 | grad norm: 0.588 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.948 | TFLOPs: 30.38 | 7: iteration 1720/ 21553 | consumed samples: 440320 | consumed tokens: 901775360 | elapsed time per iteration (s): 0.29 | learning rate: 1.978E-04 | global batch size: 256 | lm loss: 4.346478E+00 | grad norm: 0.619 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.241 | TFLOPs: 30.39 | 7: iteration 1730/ 21553 | consumed samples: 442880 | consumed tokens: 907018240 | elapsed time per iteration (s): 0.29 | learning rate: 1.978E-04 | global batch size: 256 | lm loss: 4.344378E+00 | grad norm: 0.492 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.132 | TFLOPs: 30.39 | 7: iteration 1740/ 21553 | consumed samples: 445440 | consumed tokens: 912261120 | elapsed time per iteration (s): 0.29 | learning rate: 1.977E-04 | global batch size: 256 | lm loss: 4.337615E+00 | grad norm: 0.603 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.259 | TFLOPs: 30.40 | 7: iteration 1750/ 21553 | consumed samples: 448000 | consumed tokens: 917504000 | elapsed time per iteration (s): 0.29 | learning rate: 1.977E-04 | global batch size: 256 | lm loss: 4.336272E+00 | grad norm: 0.543 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.451 | TFLOPs: 30.40 | 7: iteration 1760/ 21553 | consumed samples: 450560 | consumed tokens: 922746880 | elapsed time per iteration (s): 0.29 | learning rate: 1.977E-04 | global batch size: 256 | lm loss: 4.329754E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.877 | TFLOPs: 30.38 | 7: iteration 1770/ 21553 | consumed samples: 453120 | consumed tokens: 927989760 | elapsed time per iteration (s): 0.29 | learning rate: 1.977E-04 | global batch size: 256 | lm loss: 4.324294E+00 | grad norm: 0.503 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.994 | TFLOPs: 30.39 | 7: iteration 1780/ 21553 | consumed samples: 455680 | consumed tokens: 933232640 | elapsed time per iteration (s): 0.29 | learning rate: 1.976E-04 | global batch size: 256 | lm loss: 4.323178E+00 | grad norm: 0.507 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.905 | TFLOPs: 30.38 | 7: iteration 1790/ 21553 | consumed samples: 458240 | consumed tokens: 938475520 | elapsed time per iteration (s): 0.29 | learning rate: 1.976E-04 | global batch size: 256 | lm loss: 4.320226E+00 | grad norm: 0.655 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.018 | TFLOPs: 30.39 | 7: iteration 1800/ 21553 | consumed samples: 460800 | consumed tokens: 943718400 | elapsed time per iteration (s): 0.29 | learning rate: 1.976E-04 | global batch size: 256 | lm loss: 4.312428E+00 | grad norm: 0.583 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.139 | TFLOPs: 30.39 | 7: iteration 1810/ 21553 | consumed samples: 463360 | consumed tokens: 948961280 | elapsed time per iteration (s): 0.30 | learning rate: 1.975E-04 | global batch size: 256 | lm loss: 4.313743E+00 | grad norm: 0.476 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.707 | TFLOPs: 30.38 | 7: iteration 1820/ 21553 | consumed samples: 465920 | consumed tokens: 954204160 | elapsed time per iteration (s): 0.30 | learning rate: 1.975E-04 | global batch size: 256 | lm loss: 4.305581E+00 | grad norm: 0.549 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.556 | TFLOPs: 30.37 | 7: iteration 1830/ 21553 | consumed samples: 468480 | consumed tokens: 959447040 | elapsed time per iteration (s): 0.29 | learning rate: 1.975E-04 | global batch size: 256 | lm loss: 4.305236E+00 | grad norm: 0.494 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.819 | TFLOPs: 30.38 | 7: iteration 1840/ 21553 | consumed samples: 471040 | consumed tokens: 964689920 | elapsed time per iteration (s): 0.30 | learning rate: 1.974E-04 | global batch size: 256 | lm loss: 4.305042E+00 | grad norm: 0.530 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.650 | TFLOPs: 30.37 | 7: iteration 1850/ 21553 | consumed samples: 473600 | consumed tokens: 969932800 | elapsed time per iteration (s): 0.29 | learning rate: 1.974E-04 | global batch size: 256 | lm loss: 4.300150E+00 | grad norm: 0.471 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.151 | TFLOPs: 30.39 | 7: iteration 1860/ 21553 | consumed samples: 476160 | consumed tokens: 975175680 | elapsed time per iteration (s): 0.30 | learning rate: 1.974E-04 | global batch size: 256 | lm loss: 4.304273E+00 | grad norm: 0.545 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.443 | TFLOPs: 30.26 | 7: iteration 1870/ 21553 | consumed samples: 478720 | consumed tokens: 980418560 | elapsed time per iteration (s): 0.30 | learning rate: 1.973E-04 | global batch size: 256 | lm loss: 4.289024E+00 | grad norm: 0.465 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.554 | TFLOPs: 29.99 | 7: iteration 1880/ 21553 | consumed samples: 481280 | consumed tokens: 985661440 | elapsed time per iteration (s): 0.29 | learning rate: 1.973E-04 | global batch size: 256 | lm loss: 4.295784E+00 | grad norm: 0.544 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.847 | TFLOPs: 30.38 | 7: iteration 1890/ 21553 | consumed samples: 483840 | consumed tokens: 990904320 | elapsed time per iteration (s): 0.30 | learning rate: 1.973E-04 | global batch size: 256 | lm loss: 4.289800E+00 | grad norm: 0.461 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.578 | TFLOPs: 30.37 | 7: iteration 1900/ 21553 | consumed samples: 486400 | consumed tokens: 996147200 | elapsed time per iteration (s): 0.30 | learning rate: 1.972E-04 | global batch size: 256 | lm loss: 4.287491E+00 | grad norm: 0.560 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.817 | TFLOPs: 30.34 | 7: iteration 1910/ 21553 | consumed samples: 488960 | consumed tokens: 1001390080 | elapsed time per iteration (s): 0.29 | learning rate: 1.972E-04 | global batch size: 256 | lm loss: 4.286379E+00 | grad norm: 0.570 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.273 | TFLOPs: 30.40 | 7: iteration 1920/ 21553 | consumed samples: 491520 | consumed tokens: 1006632960 | elapsed time per iteration (s): 0.29 | learning rate: 1.972E-04 | global batch size: 256 | lm loss: 4.279427E+00 | grad norm: 0.505 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.135 | TFLOPs: 30.39 | 7: iteration 1930/ 21553 | consumed samples: 494080 | consumed tokens: 1011875840 | elapsed time per iteration (s): 0.29 | learning rate: 1.971E-04 | global batch size: 256 | lm loss: 4.282245E+00 | grad norm: 0.662 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.940 | TFLOPs: 30.38 | 7: iteration 1940/ 21553 | consumed samples: 496640 | consumed tokens: 1017118720 | elapsed time per iteration (s): 0.30 | learning rate: 1.971E-04 | global batch size: 256 | lm loss: 4.271383E+00 | grad norm: 0.467 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.565 | TFLOPs: 30.37 | 7: iteration 1950/ 21553 | consumed samples: 499200 | consumed tokens: 1022361600 | elapsed time per iteration (s): 0.30 | learning rate: 1.971E-04 | global batch size: 256 | lm loss: 4.271987E+00 | grad norm: 0.527 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.779 | TFLOPs: 30.38 | 7: iteration 1960/ 21553 | consumed samples: 501760 | consumed tokens: 1027604480 | elapsed time per iteration (s): 0.29 | learning rate: 1.970E-04 | global batch size: 256 | lm loss: 4.264587E+00 | grad norm: 0.486 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.238 | TFLOPs: 30.39 | 7: iteration 1970/ 21553 | consumed samples: 504320 | consumed tokens: 1032847360 | elapsed time per iteration (s): 0.30 | learning rate: 1.970E-04 | global batch size: 256 | lm loss: 4.262927E+00 | grad norm: 0.535 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.728 | TFLOPs: 29.99 | 7: iteration 1980/ 21553 | consumed samples: 506880 | consumed tokens: 1038090240 | elapsed time per iteration (s): 0.29 | learning rate: 1.970E-04 | global batch size: 256 | lm loss: 4.255984E+00 | grad norm: 0.488 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.100 | TFLOPs: 30.39 | 7: iteration 1990/ 21553 | consumed samples: 509440 | consumed tokens: 1043333120 | elapsed time per iteration (s): 0.30 | learning rate: 1.969E-04 | global batch size: 256 | lm loss: 4.255850E+00 | grad norm: 0.460 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.665 | TFLOPs: 30.37 | 0: [2023-03-17 00:05:55,300] [INFO] [logging.py:68:log_dist] [Rank 0] step=2000, skipped=0, lr=[0.00019691153209285103, 0.00019691153209285103, 0.00019691153209285103], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 2000/ 21553 | consumed samples: 512000 | consumed tokens: 1048576000 | elapsed time per iteration (s): 0.30 | learning rate: 1.969E-04 | global batch size: 256 | lm loss: 4.254283E+00 | grad norm: 0.526 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.626 | TFLOPs: 30.34 | 0: steps: 2000 loss: 4.2630 iter time (s): 0.307 samples/sec: 834.432 7: ----------------------------------------------------------------------------------------------- 7: validation loss at iteration 2000 | lm loss value: 4.149524E+00 | lm loss PPL: 6.340383E+01 | 7: ----------------------------------------------------------------------------------------------- 0: saving checkpoint at iteration 2000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:05:55,422] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step2000 is begin to save! 0: [2023-03-17 00:05:55,425] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:05:55,510] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:05:55,510] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:05:55,526] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:05:55,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:05:55,541] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:05:55,541] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:05:55,575] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:05:55,575] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:05:55,590] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:05:55,590] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:05:55,605] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:05:55,605] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:05:55,620] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:05:55,620] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:05:55,635] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:05:55,635] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:05:55,650] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:05:55,650] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:05:55,665] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:05:55,666] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:05:55,680] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:05:55,681] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:05:55,695] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:05:55,696] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:05:55,711] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:05:55,711] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:05:55,726] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:05:55,726] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:05:55,741] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:05:55,741] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:05:55,756] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:05:55,756] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:05:55,757] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:05:55,758] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step2000/mp_rank_00_model_states.pt 0: [2023-03-17 00:05:55,758] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:05:55,760] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:05:55,777] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:05:55,812] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:05:55,812] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 00:05:55,812] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 5: [2023-03-17 00:05:55,812] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:05:55,812] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 00:05:55,812] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 2: [2023-03-17 00:05:55,812] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:05:55,813] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:05:55,813] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 4: [2023-03-17 00:05:55,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:05:55,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:05:55,813] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 5: [2023-03-17 00:05:55,813] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 4: [2023-03-17 00:05:55,813] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 5: [2023-03-17 00:05:55,813] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 7: [2023-03-17 00:05:55,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:05:55,813] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:05:55,813] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 7: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:05:55,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 6: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 6: [2023-03-17 00:05:55,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 7: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:05:55,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 6: [2023-03-17 00:05:55,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 7: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 6: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 6: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:05:55,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 4: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:05:55,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 3: [2023-03-17 00:05:55,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 0: [2023-03-17 00:05:55,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:05:55,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 1: [2023-03-17 00:05:55,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:05:55,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 00:05:55,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 1: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:05:55,816] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 1: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:05:55,816] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 1: [2023-03-17 00:05:55,816] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:05:55,816] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 4: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:05:55,816] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:05:55,816] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 3: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:05:55,817] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 2: [2023-03-17 00:05:55,817] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 2: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 2: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:05:55,817] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 6: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:05:55,817] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 6: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:05:55,817] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 7: [2023-03-17 00:05:55,817] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 6: [2023-03-17 00:05:55,817] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:05:55,818] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 1: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:05:55,818] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 2: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 7: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:05:55,818] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:05:55,818] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 2: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 3: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:05:55,818] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 3: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:05:55,818] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 5: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:05:55,818] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 5: [2023-03-17 00:05:55,819] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:05:55,819] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,819] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:05:55,819] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 3: [2023-03-17 00:05:55,819] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,819] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 6: [2023-03-17 00:05:55,819] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:05:55,819] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:05:55,819] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 2: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:05:55,820] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:05:55,820] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 2: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 1: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:05:55,820] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 5: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:05:55,820] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 5: [2023-03-17 00:05:55,820] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 00:05:55,820] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 6: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 5: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 5: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 1: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:05:55,820] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 2: [2023-03-17 00:05:55,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:05:55,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:05:55,821] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 6: [2023-03-17 00:05:55,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:05:55,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 1: [2023-03-17 00:05:55,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:05:55,822] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 1: [2023-03-17 00:05:55,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:05:55,822] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 3: [2023-03-17 00:05:55,822] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:05:55,822] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,822] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 3: [2023-03-17 00:05:55,822] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:05:55,822] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,822] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 1: [2023-03-17 00:05:55,824] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:05:55,824] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:05:55,824] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 4: [2023-03-17 00:05:55,818] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 7: [2023-03-17 00:05:55,818] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 4: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 7: [2023-03-17 00:05:55,818] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 4: [2023-03-17 00:05:55,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:05:55,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:05:55,820] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 4: [2023-03-17 00:05:55,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 7: [2023-03-17 00:05:55,820] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 4: [2023-03-17 00:05:55,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 7: [2023-03-17 00:05:55,822] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:05:55,821] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 7: [2023-03-17 00:05:55,822] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 4: [2023-03-17 00:05:55,821] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 7: [2023-03-17 00:05:55,822] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 4: [2023-03-17 00:05:55,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:05:55,823] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:05:55,827] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:05:55,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 7: [2023-03-17 00:05:55,823] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 4: [2023-03-17 00:05:55,822] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 7: [2023-03-17 00:05:55,823] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 4: [2023-03-17 00:05:55,823] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:05:55,823] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:05:55,823] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: [2023-03-17 00:05:55,827] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:05:55,827] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: [2023-03-17 00:05:55,828] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:05:55,828] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:05:55,828] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 3: [2023-03-17 00:05:55,828] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 0: [2023-03-17 00:05:55,828] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 3: [2023-03-17 00:05:55,828] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: [2023-03-17 00:05:55,839] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step2000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:05:55,839] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step2000 is ready now! 0: successfully saved checkpoint at iteration 2000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 423.89 7: iteration 2010/ 21553 | consumed samples: 514560 | consumed tokens: 1053818880 | elapsed time per iteration (s): 0.35 | learning rate: 1.969E-04 | global batch size: 256 | lm loss: 4.251855E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 734.045 | TFLOPs: 25.70 | 7: iteration 2020/ 21553 | consumed samples: 517120 | consumed tokens: 1059061760 | elapsed time per iteration (s): 0.30 | learning rate: 1.968E-04 | global batch size: 256 | lm loss: 4.243068E+00 | grad norm: 0.654 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.807 | TFLOPs: 30.34 | 7: iteration 2030/ 21553 | consumed samples: 519680 | consumed tokens: 1064304640 | elapsed time per iteration (s): 0.30 | learning rate: 1.968E-04 | global batch size: 256 | lm loss: 4.251450E+00 | grad norm: 0.477 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.682 | TFLOPs: 30.34 | 7: iteration 2040/ 21553 | consumed samples: 522240 | consumed tokens: 1069547520 | elapsed time per iteration (s): 0.30 | learning rate: 1.968E-04 | global batch size: 256 | lm loss: 4.243074E+00 | grad norm: 0.521 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.164 | TFLOPs: 30.32 | 7: iteration 2050/ 21553 | consumed samples: 524800 | consumed tokens: 1074790400 | elapsed time per iteration (s): 0.30 | learning rate: 1.967E-04 | global batch size: 256 | lm loss: 4.235822E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.530 | TFLOPs: 30.37 | 7: iteration 2060/ 21553 | consumed samples: 527360 | consumed tokens: 1080033280 | elapsed time per iteration (s): 0.30 | learning rate: 1.967E-04 | global batch size: 256 | lm loss: 4.237513E+00 | grad norm: 0.486 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.491 | TFLOPs: 30.37 | 7: iteration 2070/ 21553 | consumed samples: 529920 | consumed tokens: 1085276160 | elapsed time per iteration (s): 0.30 | learning rate: 1.967E-04 | global batch size: 256 | lm loss: 4.232469E+00 | grad norm: 0.485 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.612 | TFLOPs: 30.37 | 7: iteration 2080/ 21553 | consumed samples: 532480 | consumed tokens: 1090519040 | elapsed time per iteration (s): 0.30 | learning rate: 1.966E-04 | global batch size: 256 | lm loss: 4.231266E+00 | grad norm: 0.531 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.694 | TFLOPs: 30.38 | 7: iteration 2090/ 21553 | consumed samples: 535040 | consumed tokens: 1095761920 | elapsed time per iteration (s): 0.29 | learning rate: 1.966E-04 | global batch size: 256 | lm loss: 4.232909E+00 | grad norm: 0.538 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.299 | TFLOPs: 30.40 | 7: iteration 2100/ 21553 | consumed samples: 537600 | consumed tokens: 1101004800 | elapsed time per iteration (s): 0.30 | learning rate: 1.966E-04 | global batch size: 256 | lm loss: 4.225435E+00 | grad norm: 0.492 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.688 | TFLOPs: 30.38 | 7: iteration 2110/ 21553 | consumed samples: 540160 | consumed tokens: 1106247680 | elapsed time per iteration (s): 0.30 | learning rate: 1.965E-04 | global batch size: 256 | lm loss: 4.223722E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.095 | TFLOPs: 30.35 | 7: iteration 2120/ 21553 | consumed samples: 542720 | consumed tokens: 1111490560 | elapsed time per iteration (s): 0.30 | learning rate: 1.965E-04 | global batch size: 256 | lm loss: 4.217228E+00 | grad norm: 0.512 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.583 | TFLOPs: 30.37 | 7: iteration 2130/ 21553 | consumed samples: 545280 | consumed tokens: 1116733440 | elapsed time per iteration (s): 0.30 | learning rate: 1.964E-04 | global batch size: 256 | lm loss: 4.214999E+00 | grad norm: 0.456 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.676 | TFLOPs: 30.37 | 7: iteration 2140/ 21553 | consumed samples: 547840 | consumed tokens: 1121976320 | elapsed time per iteration (s): 0.30 | learning rate: 1.964E-04 | global batch size: 256 | lm loss: 4.213832E+00 | grad norm: 0.516 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.539 | TFLOPs: 30.37 | 7: iteration 2150/ 21553 | consumed samples: 550400 | consumed tokens: 1127219200 | elapsed time per iteration (s): 0.30 | learning rate: 1.964E-04 | global batch size: 256 | lm loss: 4.213350E+00 | grad norm: 0.497 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.677 | TFLOPs: 30.34 | 7: iteration 2160/ 21553 | consumed samples: 552960 | consumed tokens: 1132462080 | elapsed time per iteration (s): 0.30 | learning rate: 1.963E-04 | global batch size: 256 | lm loss: 4.207357E+00 | grad norm: 0.514 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.186 | TFLOPs: 30.32 | 7: iteration 2170/ 21553 | consumed samples: 555520 | consumed tokens: 1137704960 | elapsed time per iteration (s): 0.30 | learning rate: 1.963E-04 | global batch size: 256 | lm loss: 4.199641E+00 | grad norm: 0.477 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.536 | TFLOPs: 30.34 | 7: iteration 2180/ 21553 | consumed samples: 558080 | consumed tokens: 1142947840 | elapsed time per iteration (s): 0.30 | learning rate: 1.963E-04 | global batch size: 256 | lm loss: 4.203811E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.815 | TFLOPs: 30.34 | 7: iteration 2190/ 21553 | consumed samples: 560640 | consumed tokens: 1148190720 | elapsed time per iteration (s): 0.30 | learning rate: 1.962E-04 | global batch size: 256 | lm loss: 4.198733E+00 | grad norm: 0.552 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.558 | TFLOPs: 30.34 | 7: iteration 2200/ 21553 | consumed samples: 563200 | consumed tokens: 1153433600 | elapsed time per iteration (s): 0.30 | learning rate: 1.962E-04 | global batch size: 256 | lm loss: 4.194378E+00 | grad norm: 0.468 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.602 | TFLOPs: 30.34 | 7: iteration 2210/ 21553 | consumed samples: 565760 | consumed tokens: 1158676480 | elapsed time per iteration (s): 0.30 | learning rate: 1.961E-04 | global batch size: 256 | lm loss: 4.197529E+00 | grad norm: 0.513 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.449 | TFLOPs: 30.33 | 7: iteration 2220/ 21553 | consumed samples: 568320 | consumed tokens: 1163919360 | elapsed time per iteration (s): 0.30 | learning rate: 1.961E-04 | global batch size: 256 | lm loss: 4.189754E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.996 | TFLOPs: 30.35 | 7: iteration 2230/ 21553 | consumed samples: 570880 | consumed tokens: 1169162240 | elapsed time per iteration (s): 0.30 | learning rate: 1.961E-04 | global batch size: 256 | lm loss: 4.188801E+00 | grad norm: 0.553 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.255 | TFLOPs: 30.36 | 7: iteration 2240/ 21553 | consumed samples: 573440 | consumed tokens: 1174405120 | elapsed time per iteration (s): 0.30 | learning rate: 1.960E-04 | global batch size: 256 | lm loss: 4.184661E+00 | grad norm: 0.500 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.072 | TFLOPs: 30.35 | 7: iteration 2250/ 21553 | consumed samples: 576000 | consumed tokens: 1179648000 | elapsed time per iteration (s): 0.30 | learning rate: 1.960E-04 | global batch size: 256 | lm loss: 4.181158E+00 | grad norm: 0.492 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.791 | TFLOPs: 30.34 | 7: iteration 2260/ 21553 | consumed samples: 578560 | consumed tokens: 1184890880 | elapsed time per iteration (s): 0.30 | learning rate: 1.960E-04 | global batch size: 256 | lm loss: 4.185731E+00 | grad norm: 0.628 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.755 | TFLOPs: 30.34 | 7: iteration 2270/ 21553 | consumed samples: 581120 | consumed tokens: 1190133760 | elapsed time per iteration (s): 0.30 | learning rate: 1.959E-04 | global batch size: 256 | lm loss: 4.185684E+00 | grad norm: 0.481 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.151 | TFLOPs: 30.32 | 7: iteration 2280/ 21553 | consumed samples: 583680 | consumed tokens: 1195376640 | elapsed time per iteration (s): 0.30 | learning rate: 1.959E-04 | global batch size: 256 | lm loss: 4.161427E+00 | grad norm: 0.460 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.908 | TFLOPs: 30.35 | 7: iteration 2290/ 21553 | consumed samples: 586240 | consumed tokens: 1200619520 | elapsed time per iteration (s): 0.30 | learning rate: 1.958E-04 | global batch size: 256 | lm loss: 4.173658E+00 | grad norm: 0.490 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.018 | TFLOPs: 30.35 | 7: iteration 2300/ 21553 | consumed samples: 588800 | consumed tokens: 1205862400 | elapsed time per iteration (s): 0.30 | learning rate: 1.958E-04 | global batch size: 256 | lm loss: 4.160674E+00 | grad norm: 0.456 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.906 | TFLOPs: 30.35 | 7: iteration 2310/ 21553 | consumed samples: 591360 | consumed tokens: 1211105280 | elapsed time per iteration (s): 0.30 | learning rate: 1.958E-04 | global batch size: 256 | lm loss: 4.166951E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.202 | TFLOPs: 30.32 | 7: iteration 2320/ 21553 | consumed samples: 593920 | consumed tokens: 1216348160 | elapsed time per iteration (s): 0.30 | learning rate: 1.957E-04 | global batch size: 256 | lm loss: 4.168021E+00 | grad norm: 0.727 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.067 | TFLOPs: 30.32 | 7: iteration 2330/ 21553 | consumed samples: 596480 | consumed tokens: 1221591040 | elapsed time per iteration (s): 0.30 | learning rate: 1.957E-04 | global batch size: 256 | lm loss: 4.170864E+00 | grad norm: 0.467 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.518 | TFLOPs: 30.33 | 7: iteration 2340/ 21553 | consumed samples: 599040 | consumed tokens: 1226833920 | elapsed time per iteration (s): 0.30 | learning rate: 1.956E-04 | global batch size: 256 | lm loss: 4.152526E+00 | grad norm: 0.550 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.568 | TFLOPs: 30.37 | 7: iteration 2350/ 21553 | consumed samples: 601600 | consumed tokens: 1232076800 | elapsed time per iteration (s): 0.30 | learning rate: 1.956E-04 | global batch size: 256 | lm loss: 4.156593E+00 | grad norm: 0.611 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.995 | TFLOPs: 30.07 | 7: iteration 2360/ 21553 | consumed samples: 604160 | consumed tokens: 1237319680 | elapsed time per iteration (s): 0.30 | learning rate: 1.956E-04 | global batch size: 256 | lm loss: 4.154922E+00 | grad norm: 0.496 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.160 | TFLOPs: 30.36 | 7: iteration 2370/ 21553 | consumed samples: 606720 | consumed tokens: 1242562560 | elapsed time per iteration (s): 0.30 | learning rate: 1.955E-04 | global batch size: 256 | lm loss: 4.154206E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.720 | TFLOPs: 30.38 | 7: iteration 2380/ 21553 | consumed samples: 609280 | consumed tokens: 1247805440 | elapsed time per iteration (s): 0.30 | learning rate: 1.955E-04 | global batch size: 256 | lm loss: 4.145038E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.602 | TFLOPs: 30.37 | 7: iteration 2390/ 21553 | consumed samples: 611840 | consumed tokens: 1253048320 | elapsed time per iteration (s): 0.30 | learning rate: 1.954E-04 | global batch size: 256 | lm loss: 4.152770E+00 | grad norm: 0.494 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.385 | TFLOPs: 30.36 | 7: iteration 2400/ 21553 | consumed samples: 614400 | consumed tokens: 1258291200 | elapsed time per iteration (s): 0.30 | learning rate: 1.954E-04 | global batch size: 256 | lm loss: 4.144339E+00 | grad norm: 0.563 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.548 | TFLOPs: 30.37 | 7: iteration 2410/ 21553 | consumed samples: 616960 | consumed tokens: 1263534080 | elapsed time per iteration (s): 0.30 | learning rate: 1.953E-04 | global batch size: 256 | lm loss: 4.143768E+00 | grad norm: 0.451 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.695 | TFLOPs: 30.34 | 7: iteration 2420/ 21553 | consumed samples: 619520 | consumed tokens: 1268776960 | elapsed time per iteration (s): 0.30 | learning rate: 1.953E-04 | global batch size: 256 | lm loss: 4.131081E+00 | grad norm: 0.464 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.173 | TFLOPs: 30.29 | 7: iteration 2430/ 21553 | consumed samples: 622080 | consumed tokens: 1274019840 | elapsed time per iteration (s): 0.30 | learning rate: 1.953E-04 | global batch size: 256 | lm loss: 4.135292E+00 | grad norm: 0.653 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.344 | TFLOPs: 30.33 | 7: iteration 2440/ 21553 | consumed samples: 624640 | consumed tokens: 1279262720 | elapsed time per iteration (s): 0.30 | learning rate: 1.952E-04 | global batch size: 256 | lm loss: 4.134369E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.288 | TFLOPs: 30.33 | 7: iteration 2450/ 21553 | consumed samples: 627200 | consumed tokens: 1284505600 | elapsed time per iteration (s): 0.30 | learning rate: 1.952E-04 | global batch size: 256 | lm loss: 4.137461E+00 | grad norm: 0.630 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.513 | TFLOPs: 30.33 | 7: iteration 2460/ 21553 | consumed samples: 629760 | consumed tokens: 1289748480 | elapsed time per iteration (s): 0.30 | learning rate: 1.951E-04 | global batch size: 256 | lm loss: 4.126818E+00 | grad norm: 0.522 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.467 | TFLOPs: 30.33 | 7: iteration 2470/ 21553 | consumed samples: 632320 | consumed tokens: 1294991360 | elapsed time per iteration (s): 0.30 | learning rate: 1.951E-04 | global batch size: 256 | lm loss: 4.131736E+00 | grad norm: 0.465 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.643 | TFLOPs: 30.34 | 7: iteration 2480/ 21553 | consumed samples: 634880 | consumed tokens: 1300234240 | elapsed time per iteration (s): 0.30 | learning rate: 1.950E-04 | global batch size: 256 | lm loss: 4.135974E+00 | grad norm: 0.471 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.317 | TFLOPs: 30.33 | 7: iteration 2490/ 21553 | consumed samples: 637440 | consumed tokens: 1305477120 | elapsed time per iteration (s): 0.30 | learning rate: 1.950E-04 | global batch size: 256 | lm loss: 4.125015E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.518 | TFLOPs: 30.33 | 7: iteration 2500/ 21553 | consumed samples: 640000 | consumed tokens: 1310720000 | elapsed time per iteration (s): 0.30 | learning rate: 1.950E-04 | global batch size: 256 | lm loss: 4.123751E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.537 | TFLOPs: 30.37 | 7: iteration 2510/ 21553 | consumed samples: 642560 | consumed tokens: 1315962880 | elapsed time per iteration (s): 0.30 | learning rate: 1.949E-04 | global batch size: 256 | lm loss: 4.124525E+00 | grad norm: 0.568 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.793 | TFLOPs: 30.38 | 7: iteration 2520/ 21553 | consumed samples: 645120 | consumed tokens: 1321205760 | elapsed time per iteration (s): 0.30 | learning rate: 1.949E-04 | global batch size: 256 | lm loss: 4.115203E+00 | grad norm: 0.476 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.971 | TFLOPs: 30.32 | 7: iteration 2530/ 21553 | consumed samples: 647680 | consumed tokens: 1326448640 | elapsed time per iteration (s): 0.30 | learning rate: 1.948E-04 | global batch size: 256 | lm loss: 4.120409E+00 | grad norm: 0.525 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.422 | TFLOPs: 30.37 | 7: iteration 2540/ 21553 | consumed samples: 650240 | consumed tokens: 1331691520 | elapsed time per iteration (s): 0.30 | learning rate: 1.948E-04 | global batch size: 256 | lm loss: 4.124170E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.513 | TFLOPs: 30.37 | 7: iteration 2550/ 21553 | consumed samples: 652800 | consumed tokens: 1336934400 | elapsed time per iteration (s): 0.30 | learning rate: 1.947E-04 | global batch size: 256 | lm loss: 4.116158E+00 | grad norm: 0.615 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 848.063 | TFLOPs: 29.69 | 7: iteration 2560/ 21553 | consumed samples: 655360 | consumed tokens: 1342177280 | elapsed time per iteration (s): 0.30 | learning rate: 1.947E-04 | global batch size: 256 | lm loss: 4.111029E+00 | grad norm: 0.480 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.466 | TFLOPs: 30.33 | 7: iteration 2570/ 21553 | consumed samples: 657920 | consumed tokens: 1347420160 | elapsed time per iteration (s): 0.30 | learning rate: 1.946E-04 | global batch size: 256 | lm loss: 4.107207E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 853.828 | TFLOPs: 29.89 | 7: iteration 2580/ 21553 | consumed samples: 660480 | consumed tokens: 1352663040 | elapsed time per iteration (s): 0.30 | learning rate: 1.946E-04 | global batch size: 256 | lm loss: 4.110934E+00 | grad norm: 0.482 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.580 | TFLOPs: 30.30 | 7: iteration 2590/ 21553 | consumed samples: 663040 | consumed tokens: 1357905920 | elapsed time per iteration (s): 0.30 | learning rate: 1.946E-04 | global batch size: 256 | lm loss: 4.108162E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 853.634 | TFLOPs: 29.88 | 7: iteration 2600/ 21553 | consumed samples: 665600 | consumed tokens: 1363148800 | elapsed time per iteration (s): 0.30 | learning rate: 1.945E-04 | global batch size: 256 | lm loss: 4.109272E+00 | grad norm: 0.483 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.409 | TFLOPs: 30.30 | 7: iteration 2610/ 21553 | consumed samples: 668160 | consumed tokens: 1368391680 | elapsed time per iteration (s): 0.30 | learning rate: 1.945E-04 | global batch size: 256 | lm loss: 4.101650E+00 | grad norm: 0.470 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.019 | TFLOPs: 30.32 | 7: iteration 2620/ 21553 | consumed samples: 670720 | consumed tokens: 1373634560 | elapsed time per iteration (s): 0.30 | learning rate: 1.944E-04 | global batch size: 256 | lm loss: 4.099799E+00 | grad norm: 0.450 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.064 | TFLOPs: 30.35 | 7: iteration 2630/ 21553 | consumed samples: 673280 | consumed tokens: 1378877440 | elapsed time per iteration (s): 0.30 | learning rate: 1.944E-04 | global batch size: 256 | lm loss: 4.093244E+00 | grad norm: 0.478 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.058 | TFLOPs: 30.35 | 7: iteration 2640/ 21553 | consumed samples: 675840 | consumed tokens: 1384120320 | elapsed time per iteration (s): 0.30 | learning rate: 1.943E-04 | global batch size: 256 | lm loss: 4.090623E+00 | grad norm: 0.502 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.222 | TFLOPs: 30.36 | 7: iteration 2650/ 21553 | consumed samples: 678400 | consumed tokens: 1389363200 | elapsed time per iteration (s): 0.30 | learning rate: 1.943E-04 | global batch size: 256 | lm loss: 4.099689E+00 | grad norm: 0.444 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.492 | TFLOPs: 30.37 | 7: iteration 2660/ 21553 | consumed samples: 680960 | consumed tokens: 1394606080 | elapsed time per iteration (s): 0.30 | learning rate: 1.942E-04 | global batch size: 256 | lm loss: 4.089072E+00 | grad norm: 0.501 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.321 | TFLOPs: 30.36 | 7: iteration 2670/ 21553 | consumed samples: 683520 | consumed tokens: 1399848960 | elapsed time per iteration (s): 0.30 | learning rate: 1.942E-04 | global batch size: 256 | lm loss: 4.093774E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.694 | TFLOPs: 30.38 | 7: iteration 2680/ 21553 | consumed samples: 686080 | consumed tokens: 1405091840 | elapsed time per iteration (s): 0.30 | learning rate: 1.941E-04 | global batch size: 256 | lm loss: 4.097452E+00 | grad norm: 0.631 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.305 | TFLOPs: 30.36 | 7: iteration 2690/ 21553 | consumed samples: 688640 | consumed tokens: 1410334720 | elapsed time per iteration (s): 0.30 | learning rate: 1.941E-04 | global batch size: 256 | lm loss: 4.090657E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.913 | TFLOPs: 30.35 | 7: iteration 2700/ 21553 | consumed samples: 691200 | consumed tokens: 1415577600 | elapsed time per iteration (s): 0.30 | learning rate: 1.940E-04 | global batch size: 256 | lm loss: 4.081752E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.106 | TFLOPs: 30.35 | 7: iteration 2710/ 21553 | consumed samples: 693760 | consumed tokens: 1420820480 | elapsed time per iteration (s): 0.30 | learning rate: 1.940E-04 | global batch size: 256 | lm loss: 4.090756E+00 | grad norm: 0.457 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.160 | TFLOPs: 30.36 | 7: iteration 2720/ 21553 | consumed samples: 696320 | consumed tokens: 1426063360 | elapsed time per iteration (s): 0.30 | learning rate: 1.940E-04 | global batch size: 256 | lm loss: 4.088185E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 852.406 | TFLOPs: 29.84 | 7: iteration 2730/ 21553 | consumed samples: 698880 | consumed tokens: 1431306240 | elapsed time per iteration (s): 0.30 | learning rate: 1.939E-04 | global batch size: 256 | lm loss: 4.074289E+00 | grad norm: 0.451 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.340 | TFLOPs: 30.36 | 7: iteration 2740/ 21553 | consumed samples: 701440 | consumed tokens: 1436549120 | elapsed time per iteration (s): 0.30 | learning rate: 1.939E-04 | global batch size: 256 | lm loss: 4.079214E+00 | grad norm: 0.576 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.812 | TFLOPs: 30.17 | 7: iteration 2750/ 21553 | consumed samples: 704000 | consumed tokens: 1441792000 | elapsed time per iteration (s): 0.30 | learning rate: 1.938E-04 | global batch size: 256 | lm loss: 4.083588E+00 | grad norm: 0.460 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 850.388 | TFLOPs: 29.77 | 7: iteration 2760/ 21553 | consumed samples: 706560 | consumed tokens: 1447034880 | elapsed time per iteration (s): 0.30 | learning rate: 1.938E-04 | global batch size: 256 | lm loss: 4.070572E+00 | grad norm: 0.466 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.096 | TFLOPs: 30.35 | 7: iteration 2770/ 21553 | consumed samples: 709120 | consumed tokens: 1452277760 | elapsed time per iteration (s): 0.30 | learning rate: 1.937E-04 | global batch size: 256 | lm loss: 4.070539E+00 | grad norm: 0.458 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.027 | TFLOPs: 30.35 | 7: iteration 2780/ 21553 | consumed samples: 711680 | consumed tokens: 1457520640 | elapsed time per iteration (s): 0.30 | learning rate: 1.937E-04 | global batch size: 256 | lm loss: 4.064130E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.436 | TFLOPs: 30.37 | 7: iteration 2790/ 21553 | consumed samples: 714240 | consumed tokens: 1462763520 | elapsed time per iteration (s): 0.30 | learning rate: 1.936E-04 | global batch size: 256 | lm loss: 4.061031E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.204 | TFLOPs: 30.36 | 7: iteration 2800/ 21553 | consumed samples: 716800 | consumed tokens: 1468006400 | elapsed time per iteration (s): 0.30 | learning rate: 1.936E-04 | global batch size: 256 | lm loss: 4.072615E+00 | grad norm: 0.513 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.912 | TFLOPs: 30.35 | 7: iteration 2810/ 21553 | consumed samples: 719360 | consumed tokens: 1473249280 | elapsed time per iteration (s): 0.30 | learning rate: 1.935E-04 | global batch size: 256 | lm loss: 4.058930E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.948 | TFLOPs: 30.35 | 7: iteration 2820/ 21553 | consumed samples: 721920 | consumed tokens: 1478492160 | elapsed time per iteration (s): 0.30 | learning rate: 1.935E-04 | global batch size: 256 | lm loss: 4.056604E+00 | grad norm: 0.454 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.000 | TFLOPs: 30.35 | 7: iteration 2830/ 21553 | consumed samples: 724480 | consumed tokens: 1483735040 | elapsed time per iteration (s): 0.30 | learning rate: 1.934E-04 | global batch size: 256 | lm loss: 4.062039E+00 | grad norm: 0.480 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.567 | TFLOPs: 30.34 | 7: iteration 2840/ 21553 | consumed samples: 727040 | consumed tokens: 1488977920 | elapsed time per iteration (s): 0.30 | learning rate: 1.934E-04 | global batch size: 256 | lm loss: 4.064380E+00 | grad norm: 0.553 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.453 | TFLOPs: 30.33 | 7: iteration 2850/ 21553 | consumed samples: 729600 | consumed tokens: 1494220800 | elapsed time per iteration (s): 0.30 | learning rate: 1.933E-04 | global batch size: 256 | lm loss: 4.054163E+00 | grad norm: 0.459 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.197 | TFLOPs: 30.25 | 7: iteration 2860/ 21553 | consumed samples: 732160 | consumed tokens: 1499463680 | elapsed time per iteration (s): 0.30 | learning rate: 1.933E-04 | global batch size: 256 | lm loss: 4.059766E+00 | grad norm: 0.467 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.842 | TFLOPs: 30.35 | 7: iteration 2870/ 21553 | consumed samples: 734720 | consumed tokens: 1504706560 | elapsed time per iteration (s): 0.30 | learning rate: 1.932E-04 | global batch size: 256 | lm loss: 4.051877E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.850 | TFLOPs: 30.35 | 7: iteration 2880/ 21553 | consumed samples: 737280 | consumed tokens: 1509949440 | elapsed time per iteration (s): 0.30 | learning rate: 1.932E-04 | global batch size: 256 | lm loss: 4.043431E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.673 | TFLOPs: 30.34 | 7: iteration 2890/ 21553 | consumed samples: 739840 | consumed tokens: 1515192320 | elapsed time per iteration (s): 0.30 | learning rate: 1.931E-04 | global batch size: 256 | lm loss: 4.049185E+00 | grad norm: 0.566 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.688 | TFLOPs: 30.34 | 7: iteration 2900/ 21553 | consumed samples: 742400 | consumed tokens: 1520435200 | elapsed time per iteration (s): 0.30 | learning rate: 1.931E-04 | global batch size: 256 | lm loss: 4.051639E+00 | grad norm: 0.521 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.160 | TFLOPs: 30.32 | 7: iteration 2910/ 21553 | consumed samples: 744960 | consumed tokens: 1525678080 | elapsed time per iteration (s): 0.30 | learning rate: 1.930E-04 | global batch size: 256 | lm loss: 4.047247E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.420 | TFLOPs: 30.33 | 7: iteration 2920/ 21553 | consumed samples: 747520 | consumed tokens: 1530920960 | elapsed time per iteration (s): 0.30 | learning rate: 1.930E-04 | global batch size: 256 | lm loss: 4.038710E+00 | grad norm: 0.465 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.651 | TFLOPs: 30.34 | 7: iteration 2930/ 21553 | consumed samples: 750080 | consumed tokens: 1536163840 | elapsed time per iteration (s): 0.30 | learning rate: 1.929E-04 | global batch size: 256 | lm loss: 4.041845E+00 | grad norm: 0.473 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.369 | TFLOPs: 30.33 | 7: iteration 2940/ 21553 | consumed samples: 752640 | consumed tokens: 1541406720 | elapsed time per iteration (s): 0.30 | learning rate: 1.929E-04 | global batch size: 256 | lm loss: 4.044009E+00 | grad norm: 0.463 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.504 | TFLOPs: 30.33 | 7: iteration 2950/ 21553 | consumed samples: 755200 | consumed tokens: 1546649600 | elapsed time per iteration (s): 0.30 | learning rate: 1.928E-04 | global batch size: 256 | lm loss: 4.045132E+00 | grad norm: 0.490 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.349 | TFLOPs: 30.33 | 7: iteration 2960/ 21553 | consumed samples: 757760 | consumed tokens: 1551892480 | elapsed time per iteration (s): 0.30 | learning rate: 1.928E-04 | global batch size: 256 | lm loss: 4.039650E+00 | grad norm: 0.505 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.846 | TFLOPs: 30.31 | 7: iteration 2970/ 21553 | consumed samples: 760320 | consumed tokens: 1557135360 | elapsed time per iteration (s): 0.30 | learning rate: 1.927E-04 | global batch size: 256 | lm loss: 4.040192E+00 | grad norm: 0.518 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.398 | TFLOPs: 30.30 | 7: iteration 2980/ 21553 | consumed samples: 762880 | consumed tokens: 1562378240 | elapsed time per iteration (s): 0.30 | learning rate: 1.926E-04 | global batch size: 256 | lm loss: 4.033867E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 855.170 | TFLOPs: 29.94 | 7: iteration 2990/ 21553 | consumed samples: 765440 | consumed tokens: 1567621120 | elapsed time per iteration (s): 0.30 | learning rate: 1.926E-04 | global batch size: 256 | lm loss: 4.029903E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.594 | TFLOPs: 30.34 | 7: iteration 3000/ 21553 | consumed samples: 768000 | consumed tokens: 1572864000 | elapsed time per iteration (s): 0.30 | learning rate: 1.925E-04 | global batch size: 256 | lm loss: 4.036577E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.929 | TFLOPs: 30.35 | 7: ----------------------------------------------------------------------------------------------- 7: validation loss at iteration 3000 | lm loss value: 3.989957E+00 | lm loss PPL: 5.405254E+01 | 7: ----------------------------------------------------------------------------------------------- 0: saving checkpoint at iteration 3000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:10:51,618] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step3000 is begin to save! 0: [2023-03-17 00:10:51,621] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:10:51,707] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:10:51,707] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:10:51,722] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:10:51,723] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:10:51,738] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:10:51,738] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:10:51,752] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:10:51,753] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:10:51,768] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:10:51,768] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:10:51,783] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:10:51,783] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:10:51,798] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:10:51,798] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:10:51,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:10:51,813] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:10:51,828] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:10:51,828] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:10:51,843] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:10:51,843] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:10:51,858] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:10:51,858] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:10:51,873] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:10:51,873] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:10:51,888] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:10:51,889] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:10:51,903] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:10:51,904] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:10:51,918] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:10:51,919] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:10:51,934] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:10:51,934] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:10:51,935] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:10:51,936] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step3000/mp_rank_00_model_states.pt 0: [2023-03-17 00:10:51,936] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:10:51,939] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:10:51,955] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:10:51,987] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:10:51,987] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:10:51,987] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 7: [2023-03-17 00:10:51,989] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:51,989] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:10:51,989] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 1: [2023-03-17 00:10:51,991] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:51,991] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:10:51,991] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 7: [2023-03-17 00:10:51,991] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 1: [2023-03-17 00:10:51,991] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 7: [2023-03-17 00:10:51,991] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 4: [2023-03-17 00:10:51,992] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:10:51,992] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:10:51,992] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 1: [2023-03-17 00:10:51,992] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 4: [2023-03-17 00:10:51,992] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 1: [2023-03-17 00:10:51,992] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 1: [2023-03-17 00:10:51,992] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:10:51,992] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 00:10:51,992] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 4: [2023-03-17 00:10:51,992] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:10:51,992] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 00:10:51,992] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 6: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:10:51,993] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 00:10:51,993] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 6: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 3: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:10:51,993] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 2: [2023-03-17 00:10:51,993] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 6: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:10:51,993] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 7: [2023-03-17 00:10:51,993] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:51,994] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 0: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:10:51,994] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 6: [2023-03-17 00:10:51,994] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 0: [2023-03-17 00:10:51,994] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 6: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 0: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 4: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:10:51,994] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 3: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:10:51,994] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 4: [2023-03-17 00:10:51,995] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:10:51,995] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:10:51,995] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 3: [2023-03-17 00:10:51,995] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 1: [2023-03-17 00:10:51,995] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:10:51,995] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 3: [2023-03-17 00:10:51,995] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 1: [2023-03-17 00:10:51,995] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 00:10:51,995] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 0: [2023-03-17 00:10:51,995] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:10:51,996] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 3: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:10:51,996] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 00:10:51,996] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 7: [2023-03-17 00:10:51,996] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 3: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 7: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 4: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:10:51,996] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 6: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:10:51,996] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 7: [2023-03-17 00:10:51,996] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 1: [2023-03-17 00:10:51,996] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 7: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 1: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 6: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 6: [2023-03-17 00:10:51,996] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:10:51,997] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 5: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:10:51,997] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 7: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:51,997] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 4: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 4: [2023-03-17 00:10:51,997] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 1: [2023-03-17 00:10:51,998] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:10:51,998] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:10:51,999] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:10:51,999] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:10:51,999] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 1: [2023-03-17 00:10:51,999] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 6: [2023-03-17 00:10:51,999] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:10:51,999] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 00:10:51,999] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 5: [2023-03-17 00:10:51,999] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:10:51,999] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:10:51,999] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:10:51,999] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 00:10:51,999] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 5: [2023-03-17 00:10:51,999] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 5: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 3: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 3: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 0: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 4: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 3: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 0: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 4: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 5: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 6: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 7: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 5: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 6: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 5: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 5: [2023-03-17 00:10:52,001] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:10:52,001] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 00:10:52,001] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 3: [2023-03-17 00:10:52,001] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:10:52,001] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 00:10:52,001] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 4: [2023-03-17 00:10:52,001] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:10:52,001] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 00:10:52,001] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 7: [2023-03-17 00:10:52,004] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:10:52,004] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:10:52,004] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 2: [2023-03-17 00:10:51,994] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 2: [2023-03-17 00:10:51,994] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:10:51,994] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 2: [2023-03-17 00:10:51,995] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:10:51,995] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:10:51,995] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 2: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:10:51,997] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 2: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:10:51,997] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:10:51,997] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 2: [2023-03-17 00:10:51,998] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:10:51,998] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:10:51,998] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 2: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:10:52,000] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:10:52,000] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 5: [2023-03-17 00:10:52,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:10:52,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 00:10:52,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 0: [2023-03-17 00:10:52,015] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:10:52,015] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 00:10:52,015] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 0: [2023-03-17 00:10:52,015] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:10:52,015] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:10:52,015] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:10:52,015] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 00:10:52,015] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 0: [2023-03-17 00:10:52,015] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 0: [2023-03-17 00:10:52,017] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step3000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:10:52,017] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step3000 is ready now! 0: successfully saved checkpoint at iteration 3000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 403.74 7: iteration 3010/ 21553 | consumed samples: 770560 | consumed tokens: 1578106880 | elapsed time per iteration (s): 0.35 | learning rate: 1.925E-04 | global batch size: 256 | lm loss: 4.024205E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 741.421 | TFLOPs: 25.96 | 7: iteration 3020/ 21553 | consumed samples: 773120 | consumed tokens: 1583349760 | elapsed time per iteration (s): 0.30 | learning rate: 1.924E-04 | global batch size: 256 | lm loss: 4.023021E+00 | grad norm: 0.488 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.410 | TFLOPs: 30.33 | 7: iteration 3030/ 21553 | consumed samples: 775680 | consumed tokens: 1588592640 | elapsed time per iteration (s): 0.30 | learning rate: 1.924E-04 | global batch size: 256 | lm loss: 4.023437E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.422 | TFLOPs: 30.30 | 7: iteration 3040/ 21553 | consumed samples: 778240 | consumed tokens: 1593835520 | elapsed time per iteration (s): 0.30 | learning rate: 1.923E-04 | global batch size: 256 | lm loss: 4.020893E+00 | grad norm: 0.505 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.570 | TFLOPs: 30.27 | 7: iteration 3050/ 21553 | consumed samples: 780800 | consumed tokens: 1599078400 | elapsed time per iteration (s): 0.30 | learning rate: 1.923E-04 | global batch size: 256 | lm loss: 4.022620E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.294 | TFLOPs: 30.36 | 7: iteration 3060/ 21553 | consumed samples: 783360 | consumed tokens: 1604321280 | elapsed time per iteration (s): 0.30 | learning rate: 1.922E-04 | global batch size: 256 | lm loss: 4.014522E+00 | grad norm: 0.555 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 850.080 | TFLOPs: 29.76 | 7: iteration 3070/ 21553 | consumed samples: 785920 | consumed tokens: 1609564160 | elapsed time per iteration (s): 0.30 | learning rate: 1.922E-04 | global batch size: 256 | lm loss: 4.023281E+00 | grad norm: 0.523 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.538 | TFLOPs: 30.37 | 7: iteration 3080/ 21553 | consumed samples: 788480 | consumed tokens: 1614807040 | elapsed time per iteration (s): 0.30 | learning rate: 1.921E-04 | global batch size: 256 | lm loss: 4.013620E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.313 | TFLOPs: 30.36 | 7: iteration 3090/ 21553 | consumed samples: 791040 | consumed tokens: 1620049920 | elapsed time per iteration (s): 0.30 | learning rate: 1.921E-04 | global batch size: 256 | lm loss: 4.022199E+00 | grad norm: 0.549 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.825 | TFLOPs: 30.31 | 7: iteration 3100/ 21553 | consumed samples: 793600 | consumed tokens: 1625292800 | elapsed time per iteration (s): 0.30 | learning rate: 1.920E-04 | global batch size: 256 | lm loss: 4.023568E+00 | grad norm: 0.487 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.793 | TFLOPs: 30.31 | 7: iteration 3110/ 21553 | consumed samples: 796160 | consumed tokens: 1630535680 | elapsed time per iteration (s): 0.30 | learning rate: 1.920E-04 | global batch size: 256 | lm loss: 4.014246E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.486 | TFLOPs: 30.23 | 7: iteration 3120/ 21553 | consumed samples: 798720 | consumed tokens: 1635778560 | elapsed time per iteration (s): 0.30 | learning rate: 1.919E-04 | global batch size: 256 | lm loss: 4.013161E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.966 | TFLOPs: 30.32 | 7: iteration 3130/ 21553 | consumed samples: 801280 | consumed tokens: 1641021440 | elapsed time per iteration (s): 0.30 | learning rate: 1.918E-04 | global batch size: 256 | lm loss: 4.011105E+00 | grad norm: 0.479 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.084 | TFLOPs: 30.32 | 7: iteration 3140/ 21553 | consumed samples: 803840 | consumed tokens: 1646264320 | elapsed time per iteration (s): 0.30 | learning rate: 1.918E-04 | global batch size: 256 | lm loss: 4.012011E+00 | grad norm: 0.519 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.704 | TFLOPs: 30.31 | 7: iteration 3150/ 21553 | consumed samples: 806400 | consumed tokens: 1651507200 | elapsed time per iteration (s): 0.30 | learning rate: 1.917E-04 | global batch size: 256 | lm loss: 4.014001E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.318 | TFLOPs: 30.33 | 7: iteration 3160/ 21553 | consumed samples: 808960 | consumed tokens: 1656750080 | elapsed time per iteration (s): 0.30 | learning rate: 1.917E-04 | global batch size: 256 | lm loss: 4.014843E+00 | grad norm: 0.480 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.400 | TFLOPs: 30.33 | 7: iteration 3170/ 21553 | consumed samples: 811520 | consumed tokens: 1661992960 | elapsed time per iteration (s): 0.30 | learning rate: 1.916E-04 | global batch size: 256 | lm loss: 3.994975E+00 | grad norm: 0.444 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.189 | TFLOPs: 30.32 | 7: iteration 3180/ 21553 | consumed samples: 814080 | consumed tokens: 1667235840 | elapsed time per iteration (s): 0.30 | learning rate: 1.916E-04 | global batch size: 256 | lm loss: 3.994479E+00 | grad norm: 0.393 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.829 | TFLOPs: 30.31 | 7: iteration 3190/ 21553 | consumed samples: 816640 | consumed tokens: 1672478720 | elapsed time per iteration (s): 0.30 | learning rate: 1.915E-04 | global batch size: 256 | lm loss: 3.993795E+00 | grad norm: 0.450 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.174 | TFLOPs: 30.32 | 7: iteration 3200/ 21553 | consumed samples: 819200 | consumed tokens: 1677721600 | elapsed time per iteration (s): 0.30 | learning rate: 1.915E-04 | global batch size: 256 | lm loss: 4.000530E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.131 | TFLOPs: 30.22 | 7: iteration 3210/ 21553 | consumed samples: 821760 | consumed tokens: 1682964480 | elapsed time per iteration (s): 0.30 | learning rate: 1.914E-04 | global batch size: 256 | lm loss: 3.996465E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.826 | TFLOPs: 30.24 | 7: iteration 3220/ 21553 | consumed samples: 824320 | consumed tokens: 1688207360 | elapsed time per iteration (s): 0.30 | learning rate: 1.913E-04 | global batch size: 256 | lm loss: 3.995078E+00 | grad norm: 0.488 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.910 | TFLOPs: 30.24 | 7: iteration 3230/ 21553 | consumed samples: 826880 | consumed tokens: 1693450240 | elapsed time per iteration (s): 0.30 | learning rate: 1.913E-04 | global batch size: 256 | lm loss: 3.994704E+00 | grad norm: 0.450 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.091 | TFLOPs: 30.25 | 7: iteration 3240/ 21553 | consumed samples: 829440 | consumed tokens: 1698693120 | elapsed time per iteration (s): 0.30 | learning rate: 1.912E-04 | global batch size: 256 | lm loss: 3.991736E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.598 | TFLOPs: 30.23 | 7: iteration 3250/ 21553 | consumed samples: 832000 | consumed tokens: 1703936000 | elapsed time per iteration (s): 0.30 | learning rate: 1.912E-04 | global batch size: 256 | lm loss: 3.995694E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.340 | TFLOPs: 30.26 | 7: iteration 3260/ 21553 | consumed samples: 834560 | consumed tokens: 1709178880 | elapsed time per iteration (s): 0.30 | learning rate: 1.911E-04 | global batch size: 256 | lm loss: 3.993081E+00 | grad norm: 0.460 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.892 | TFLOPs: 30.24 | 7: iteration 3270/ 21553 | consumed samples: 837120 | consumed tokens: 1714421760 | elapsed time per iteration (s): 0.30 | learning rate: 1.911E-04 | global batch size: 256 | lm loss: 3.990859E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.149 | TFLOPs: 30.22 | 7: iteration 3280/ 21553 | consumed samples: 839680 | consumed tokens: 1719664640 | elapsed time per iteration (s): 0.30 | learning rate: 1.910E-04 | global batch size: 256 | lm loss: 3.980405E+00 | grad norm: 0.479 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.609 | TFLOPs: 30.23 | 7: iteration 3290/ 21553 | consumed samples: 842240 | consumed tokens: 1724907520 | elapsed time per iteration (s): 0.30 | learning rate: 1.909E-04 | global batch size: 256 | lm loss: 3.986541E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.092 | TFLOPs: 30.21 | 7: iteration 3300/ 21553 | consumed samples: 844800 | consumed tokens: 1730150400 | elapsed time per iteration (s): 0.30 | learning rate: 1.909E-04 | global batch size: 256 | lm loss: 3.976037E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.630 | TFLOPs: 30.20 | 7: iteration 3310/ 21553 | consumed samples: 847360 | consumed tokens: 1735393280 | elapsed time per iteration (s): 0.30 | learning rate: 1.908E-04 | global batch size: 256 | lm loss: 3.988315E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.526 | TFLOPs: 30.23 | 7: iteration 3320/ 21553 | consumed samples: 849920 | consumed tokens: 1740636160 | elapsed time per iteration (s): 0.30 | learning rate: 1.908E-04 | global batch size: 256 | lm loss: 3.981139E+00 | grad norm: 0.444 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.732 | TFLOPs: 30.24 | 7: iteration 3330/ 21553 | consumed samples: 852480 | consumed tokens: 1745879040 | elapsed time per iteration (s): 0.30 | learning rate: 1.907E-04 | global batch size: 256 | lm loss: 3.977179E+00 | grad norm: 0.484 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.690 | TFLOPs: 30.24 | 7: iteration 3340/ 21553 | consumed samples: 855040 | consumed tokens: 1751121920 | elapsed time per iteration (s): 0.30 | learning rate: 1.906E-04 | global batch size: 256 | lm loss: 3.973022E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.588 | TFLOPs: 30.23 | 7: iteration 3350/ 21553 | consumed samples: 857600 | consumed tokens: 1756364800 | elapsed time per iteration (s): 0.30 | learning rate: 1.906E-04 | global batch size: 256 | lm loss: 3.974792E+00 | grad norm: 0.528 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.570 | TFLOPs: 30.23 | 7: iteration 3360/ 21553 | consumed samples: 860160 | consumed tokens: 1761607680 | elapsed time per iteration (s): 0.30 | learning rate: 1.905E-04 | global batch size: 256 | lm loss: 3.967100E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.139 | TFLOPs: 30.25 | 7: iteration 3370/ 21553 | consumed samples: 862720 | consumed tokens: 1766850560 | elapsed time per iteration (s): 0.30 | learning rate: 1.905E-04 | global batch size: 256 | lm loss: 3.971087E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.902 | TFLOPs: 30.24 | 7: iteration 3380/ 21553 | consumed samples: 865280 | consumed tokens: 1772093440 | elapsed time per iteration (s): 0.30 | learning rate: 1.904E-04 | global batch size: 256 | lm loss: 3.973510E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.495 | TFLOPs: 30.26 | 7: iteration 3390/ 21553 | consumed samples: 867840 | consumed tokens: 1777336320 | elapsed time per iteration (s): 0.30 | learning rate: 1.903E-04 | global batch size: 256 | lm loss: 3.982752E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.131 | TFLOPs: 30.25 | 7: iteration 3400/ 21553 | consumed samples: 870400 | consumed tokens: 1782579200 | elapsed time per iteration (s): 0.30 | learning rate: 1.903E-04 | global batch size: 256 | lm loss: 3.973845E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.842 | TFLOPs: 30.24 | 7: iteration 3410/ 21553 | consumed samples: 872960 | consumed tokens: 1787822080 | elapsed time per iteration (s): 0.30 | learning rate: 1.902E-04 | global batch size: 256 | lm loss: 3.968925E+00 | grad norm: 0.465 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.361 | TFLOPs: 30.26 | 7: iteration 3420/ 21553 | consumed samples: 875520 | consumed tokens: 1793064960 | elapsed time per iteration (s): 0.30 | learning rate: 1.902E-04 | global batch size: 256 | lm loss: 3.959633E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.441 | TFLOPs: 30.23 | 7: iteration 3430/ 21553 | consumed samples: 878080 | consumed tokens: 1798307840 | elapsed time per iteration (s): 0.30 | learning rate: 1.901E-04 | global batch size: 256 | lm loss: 3.962138E+00 | grad norm: 0.456 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.181 | TFLOPs: 30.25 | 7: iteration 3440/ 21553 | consumed samples: 880640 | consumed tokens: 1803550720 | elapsed time per iteration (s): 0.30 | learning rate: 1.900E-04 | global batch size: 256 | lm loss: 3.971517E+00 | grad norm: 0.491 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.587 | TFLOPs: 30.23 | 7: iteration 3450/ 21553 | consumed samples: 883200 | consumed tokens: 1808793600 | elapsed time per iteration (s): 0.30 | learning rate: 1.900E-04 | global batch size: 256 | lm loss: 3.960847E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.996 | TFLOPs: 30.25 | 7: iteration 3460/ 21553 | consumed samples: 885760 | consumed tokens: 1814036480 | elapsed time per iteration (s): 0.30 | learning rate: 1.899E-04 | global batch size: 256 | lm loss: 3.961460E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.617 | TFLOPs: 30.23 | 7: iteration 3470/ 21553 | consumed samples: 888320 | consumed tokens: 1819279360 | elapsed time per iteration (s): 0.30 | learning rate: 1.899E-04 | global batch size: 256 | lm loss: 3.955867E+00 | grad norm: 0.461 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.980 | TFLOPs: 30.25 | 7: iteration 3480/ 21553 | consumed samples: 890880 | consumed tokens: 1824522240 | elapsed time per iteration (s): 0.30 | learning rate: 1.898E-04 | global batch size: 256 | lm loss: 3.953837E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.535 | TFLOPs: 30.23 | 7: iteration 3490/ 21553 | consumed samples: 893440 | consumed tokens: 1829765120 | elapsed time per iteration (s): 0.30 | learning rate: 1.897E-04 | global batch size: 256 | lm loss: 3.953881E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.323 | TFLOPs: 30.22 | 7: iteration 3500/ 21553 | consumed samples: 896000 | consumed tokens: 1835008000 | elapsed time per iteration (s): 0.30 | learning rate: 1.897E-04 | global batch size: 256 | lm loss: 3.953408E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.971 | TFLOPs: 30.25 | 7: iteration 3510/ 21553 | consumed samples: 898560 | consumed tokens: 1840250880 | elapsed time per iteration (s): 0.30 | learning rate: 1.896E-04 | global batch size: 256 | lm loss: 3.959333E+00 | grad norm: 0.486 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.813 | TFLOPs: 30.20 | 7: iteration 3520/ 21553 | consumed samples: 901120 | consumed tokens: 1845493760 | elapsed time per iteration (s): 0.30 | learning rate: 1.896E-04 | global batch size: 256 | lm loss: 3.956294E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.598 | TFLOPs: 30.23 | 7: iteration 3530/ 21553 | consumed samples: 903680 | consumed tokens: 1850736640 | elapsed time per iteration (s): 0.30 | learning rate: 1.895E-04 | global batch size: 256 | lm loss: 3.956963E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.939 | TFLOPs: 30.28 | 7: iteration 3540/ 21553 | consumed samples: 906240 | consumed tokens: 1855979520 | elapsed time per iteration (s): 0.30 | learning rate: 1.894E-04 | global batch size: 256 | lm loss: 3.950253E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.474 | TFLOPs: 30.33 | 7: iteration 3550/ 21553 | consumed samples: 908800 | consumed tokens: 1861222400 | elapsed time per iteration (s): 0.30 | learning rate: 1.894E-04 | global batch size: 256 | lm loss: 3.949940E+00 | grad norm: 0.536 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.948 | TFLOPs: 30.35 | 7: iteration 3560/ 21553 | consumed samples: 911360 | consumed tokens: 1866465280 | elapsed time per iteration (s): 0.30 | learning rate: 1.893E-04 | global batch size: 256 | lm loss: 3.953325E+00 | grad norm: 0.521 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.593 | TFLOPs: 30.34 | 7: iteration 3570/ 21553 | consumed samples: 913920 | consumed tokens: 1871708160 | elapsed time per iteration (s): 0.30 | learning rate: 1.892E-04 | global batch size: 256 | lm loss: 3.954466E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.575 | TFLOPs: 30.34 | 7: iteration 3580/ 21553 | consumed samples: 916480 | consumed tokens: 1876951040 | elapsed time per iteration (s): 0.30 | learning rate: 1.892E-04 | global batch size: 256 | lm loss: 3.940896E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.430 | TFLOPs: 30.33 | 7: iteration 3590/ 21553 | consumed samples: 919040 | consumed tokens: 1882193920 | elapsed time per iteration (s): 0.30 | learning rate: 1.891E-04 | global batch size: 256 | lm loss: 3.943795E+00 | grad norm: 0.508 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.956 | TFLOPs: 30.07 | 7: iteration 3600/ 21553 | consumed samples: 921600 | consumed tokens: 1887436800 | elapsed time per iteration (s): 0.30 | learning rate: 1.891E-04 | global batch size: 256 | lm loss: 3.944865E+00 | grad norm: 0.464 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.811 | TFLOPs: 30.34 | 7: iteration 3610/ 21553 | consumed samples: 924160 | consumed tokens: 1892679680 | elapsed time per iteration (s): 0.30 | learning rate: 1.890E-04 | global batch size: 256 | lm loss: 3.944619E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.961 | TFLOPs: 30.35 | 7: iteration 3620/ 21553 | consumed samples: 926720 | consumed tokens: 1897922560 | elapsed time per iteration (s): 0.30 | learning rate: 1.889E-04 | global batch size: 256 | lm loss: 3.943105E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.892 | TFLOPs: 30.35 | 7: iteration 3630/ 21553 | consumed samples: 929280 | consumed tokens: 1903165440 | elapsed time per iteration (s): 0.30 | learning rate: 1.889E-04 | global batch size: 256 | lm loss: 3.944679E+00 | grad norm: 0.476 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.072 | TFLOPs: 30.35 | 7: iteration 3640/ 21553 | consumed samples: 931840 | consumed tokens: 1908408320 | elapsed time per iteration (s): 0.30 | learning rate: 1.888E-04 | global batch size: 256 | lm loss: 3.934251E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.798 | TFLOPs: 30.34 | 7: iteration 3650/ 21553 | consumed samples: 934400 | consumed tokens: 1913651200 | elapsed time per iteration (s): 0.30 | learning rate: 1.887E-04 | global batch size: 256 | lm loss: 3.937986E+00 | grad norm: 0.382 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.780 | TFLOPs: 30.34 | 7: iteration 3660/ 21553 | consumed samples: 936960 | consumed tokens: 1918894080 | elapsed time per iteration (s): 0.30 | learning rate: 1.887E-04 | global batch size: 256 | lm loss: 3.929842E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.590 | TFLOPs: 30.34 | 7: iteration 3670/ 21553 | consumed samples: 939520 | consumed tokens: 1924136960 | elapsed time per iteration (s): 0.30 | learning rate: 1.886E-04 | global batch size: 256 | lm loss: 3.932153E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.524 | TFLOPs: 30.33 | 7: iteration 3680/ 21553 | consumed samples: 942080 | consumed tokens: 1929379840 | elapsed time per iteration (s): 0.30 | learning rate: 1.885E-04 | global batch size: 256 | lm loss: 3.933698E+00 | grad norm: 0.468 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.768 | TFLOPs: 30.31 | 7: iteration 3690/ 21553 | consumed samples: 944640 | consumed tokens: 1934622720 | elapsed time per iteration (s): 0.30 | learning rate: 1.885E-04 | global batch size: 256 | lm loss: 3.926103E+00 | grad norm: 0.468 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.956 | TFLOPs: 30.03 | 7: iteration 3700/ 21553 | consumed samples: 947200 | consumed tokens: 1939865600 | elapsed time per iteration (s): 0.30 | learning rate: 1.884E-04 | global batch size: 256 | lm loss: 3.932782E+00 | grad norm: 0.485 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.520 | TFLOPs: 30.33 | 7: iteration 3710/ 21553 | consumed samples: 949760 | consumed tokens: 1945108480 | elapsed time per iteration (s): 0.30 | learning rate: 1.883E-04 | global batch size: 256 | lm loss: 3.928799E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.217 | TFLOPs: 30.36 | 7: iteration 3720/ 21553 | consumed samples: 952320 | consumed tokens: 1950351360 | elapsed time per iteration (s): 0.30 | learning rate: 1.883E-04 | global batch size: 256 | lm loss: 3.930278E+00 | grad norm: 0.489 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.518 | TFLOPs: 30.37 | 7: iteration 3730/ 21553 | consumed samples: 954880 | consumed tokens: 1955594240 | elapsed time per iteration (s): 0.30 | learning rate: 1.882E-04 | global batch size: 256 | lm loss: 3.930396E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.683 | TFLOPs: 30.38 | 7: iteration 3740/ 21553 | consumed samples: 957440 | consumed tokens: 1960837120 | elapsed time per iteration (s): 0.30 | learning rate: 1.882E-04 | global batch size: 256 | lm loss: 3.926075E+00 | grad norm: 0.528 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.981 | TFLOPs: 30.35 | 7: iteration 3750/ 21553 | consumed samples: 960000 | consumed tokens: 1966080000 | elapsed time per iteration (s): 0.30 | learning rate: 1.881E-04 | global batch size: 256 | lm loss: 3.922397E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.642 | TFLOPs: 30.34 | 7: iteration 3760/ 21553 | consumed samples: 962560 | consumed tokens: 1971322880 | elapsed time per iteration (s): 0.30 | learning rate: 1.880E-04 | global batch size: 256 | lm loss: 3.919041E+00 | grad norm: 0.469 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.179 | TFLOPs: 30.36 | 7: iteration 3770/ 21553 | consumed samples: 965120 | consumed tokens: 1976565760 | elapsed time per iteration (s): 0.30 | learning rate: 1.880E-04 | global batch size: 256 | lm loss: 3.923043E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.025 | TFLOPs: 30.35 | 7: iteration 3780/ 21553 | consumed samples: 967680 | consumed tokens: 1981808640 | elapsed time per iteration (s): 0.30 | learning rate: 1.879E-04 | global batch size: 256 | lm loss: 3.924323E+00 | grad norm: 0.470 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.211 | TFLOPs: 30.36 | 7: iteration 3790/ 21553 | consumed samples: 970240 | consumed tokens: 1987051520 | elapsed time per iteration (s): 0.30 | learning rate: 1.878E-04 | global batch size: 256 | lm loss: 3.926200E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.448 | TFLOPs: 30.37 | 7: iteration 3800/ 21553 | consumed samples: 972800 | consumed tokens: 1992294400 | elapsed time per iteration (s): 0.30 | learning rate: 1.878E-04 | global batch size: 256 | lm loss: 3.915554E+00 | grad norm: 0.471 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.384 | TFLOPs: 30.36 | 7: iteration 3810/ 21553 | consumed samples: 975360 | consumed tokens: 1997537280 | elapsed time per iteration (s): 0.30 | learning rate: 1.877E-04 | global batch size: 256 | lm loss: 3.919874E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.442 | TFLOPs: 30.37 | 7: iteration 3820/ 21553 | consumed samples: 977920 | consumed tokens: 2002780160 | elapsed time per iteration (s): 0.30 | learning rate: 1.876E-04 | global batch size: 256 | lm loss: 3.915237E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.261 | TFLOPs: 30.36 | 7: iteration 3830/ 21553 | consumed samples: 980480 | consumed tokens: 2008023040 | elapsed time per iteration (s): 0.30 | learning rate: 1.876E-04 | global batch size: 256 | lm loss: 3.917214E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.074 | TFLOPs: 30.35 | 7: iteration 3840/ 21553 | consumed samples: 983040 | consumed tokens: 2013265920 | elapsed time per iteration (s): 0.30 | learning rate: 1.875E-04 | global batch size: 256 | lm loss: 3.908748E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.469 | TFLOPs: 30.37 | 7: iteration 3850/ 21553 | consumed samples: 985600 | consumed tokens: 2018508800 | elapsed time per iteration (s): 0.30 | learning rate: 1.874E-04 | global batch size: 256 | lm loss: 3.917981E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.111 | TFLOPs: 30.36 | 7: iteration 3860/ 21553 | consumed samples: 988160 | consumed tokens: 2023751680 | elapsed time per iteration (s): 0.30 | learning rate: 1.874E-04 | global batch size: 256 | lm loss: 3.910250E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.784 | TFLOPs: 30.34 | 7: iteration 3870/ 21553 | consumed samples: 990720 | consumed tokens: 2028994560 | elapsed time per iteration (s): 0.30 | learning rate: 1.873E-04 | global batch size: 256 | lm loss: 3.905227E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.342 | TFLOPs: 30.36 | 7: iteration 3880/ 21553 | consumed samples: 993280 | consumed tokens: 2034237440 | elapsed time per iteration (s): 0.30 | learning rate: 1.872E-04 | global batch size: 256 | lm loss: 3.907873E+00 | grad norm: 0.444 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.375 | TFLOPs: 30.36 | 7: iteration 3890/ 21553 | consumed samples: 995840 | consumed tokens: 2039480320 | elapsed time per iteration (s): 0.31 | learning rate: 1.871E-04 | global batch size: 256 | lm loss: 3.914665E+00 | grad norm: 0.534 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 837.026 | TFLOPs: 29.30 | 7: iteration 3900/ 21553 | consumed samples: 998400 | consumed tokens: 2044723200 | elapsed time per iteration (s): 0.30 | learning rate: 1.871E-04 | global batch size: 256 | lm loss: 3.912171E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.082 | TFLOPs: 30.35 | 7: iteration 3910/ 21553 | consumed samples: 1000960 | consumed tokens: 2049966080 | elapsed time per iteration (s): 0.30 | learning rate: 1.870E-04 | global batch size: 256 | lm loss: 3.906307E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.843 | TFLOPs: 30.35 | 7: iteration 3920/ 21553 | consumed samples: 1003520 | consumed tokens: 2055208960 | elapsed time per iteration (s): 0.30 | learning rate: 1.869E-04 | global batch size: 256 | lm loss: 3.903585E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.702 | TFLOPs: 30.34 | 7: iteration 3930/ 21553 | consumed samples: 1006080 | consumed tokens: 2060451840 | elapsed time per iteration (s): 0.30 | learning rate: 1.869E-04 | global batch size: 256 | lm loss: 3.903262E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.628 | TFLOPs: 30.34 | 7: iteration 3940/ 21553 | consumed samples: 1008640 | consumed tokens: 2065694720 | elapsed time per iteration (s): 0.30 | learning rate: 1.868E-04 | global batch size: 256 | lm loss: 3.901995E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.576 | TFLOPs: 30.34 | 7: iteration 3950/ 21553 | consumed samples: 1011200 | consumed tokens: 2070937600 | elapsed time per iteration (s): 0.30 | learning rate: 1.867E-04 | global batch size: 256 | lm loss: 3.893162E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.895 | TFLOPs: 30.35 | 7: iteration 3960/ 21553 | consumed samples: 1013760 | consumed tokens: 2076180480 | elapsed time per iteration (s): 0.30 | learning rate: 1.867E-04 | global batch size: 256 | lm loss: 3.894120E+00 | grad norm: 0.470 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.993 | TFLOPs: 30.35 | 7: iteration 3970/ 21553 | consumed samples: 1016320 | consumed tokens: 2081423360 | elapsed time per iteration (s): 0.30 | learning rate: 1.866E-04 | global batch size: 256 | lm loss: 3.898368E+00 | grad norm: 0.444 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.010 | TFLOPs: 30.35 | 7: iteration 3980/ 21553 | consumed samples: 1018880 | consumed tokens: 2086666240 | elapsed time per iteration (s): 0.30 | learning rate: 1.865E-04 | global batch size: 256 | lm loss: 3.898537E+00 | grad norm: 0.492 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.976 | TFLOPs: 30.35 | 7: iteration 3990/ 21553 | consumed samples: 1021440 | consumed tokens: 2091909120 | elapsed time per iteration (s): 0.30 | learning rate: 1.865E-04 | global batch size: 256 | lm loss: 3.898239E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.329 | TFLOPs: 30.36 | 0: [2023-03-17 00:15:47,954] [INFO] [logging.py:68:log_dist] [Rank 0] step=4000, skipped=0, lr=[0.00018638652011758862, 0.00018638652011758862, 0.00018638652011758862], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 4000/ 21553 | consumed samples: 1024000 | consumed tokens: 2097152000 | elapsed time per iteration (s): 0.30 | learning rate: 1.864E-04 | global batch size: 256 | lm loss: 3.889269E+00 | grad norm: 0.380 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.440 | TFLOPs: 30.37 | 0: steps: 4000 loss: 3.8546 iter time (s): 0.294 samples/sec: 871.474 7: ----------------------------------------------------------------------------------------------- 7: validation loss at iteration 4000 | lm loss value: 3.947387E+00 | lm loss PPL: 5.179981E+01 | 7: ----------------------------------------------------------------------------------------------- 0: saving checkpoint at iteration 4000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:15:48,076] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step4000 is begin to save! 0: [2023-03-17 00:15:48,079] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:15:48,164] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:15:48,164] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:15:48,180] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:15:48,180] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:15:48,195] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:15:48,195] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:15:48,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:15:48,210] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:15:48,225] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:15:48,225] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:15:48,241] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:15:48,242] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:15:48,257] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:15:48,257] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:15:48,272] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:15:48,272] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:15:48,287] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:15:48,287] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:15:48,302] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:15:48,302] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:15:48,317] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:15:48,317] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:15:48,332] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:15:48,332] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:15:48,347] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:15:48,347] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:15:48,362] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:15:48,362] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:15:48,377] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:15:48,377] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:15:48,392] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:15:48,392] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:15:48,393] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:15:48,394] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step4000/mp_rank_00_model_states.pt 0: [2023-03-17 00:15:48,394] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:15:48,396] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:15:48,413] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:15:48,446] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,446] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,446] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 00:15:48,446] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 3: [2023-03-17 00:15:48,448] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,448] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,448] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,448] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 6: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 5: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 0: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 3: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 0: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 6: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 6: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 3: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 5: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 2: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 0: [2023-03-17 00:15:48,450] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:15:48,450] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 00:15:48,450] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 0: [2023-03-17 00:15:48,450] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:15:48,450] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 00:15:48,450] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 7: [2023-03-17 00:15:48,448] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,448] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 00:15:48,448] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 7: [2023-03-17 00:15:48,448] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,448] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 3: [2023-03-17 00:15:48,451] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,451] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 00:15:48,451] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 0: [2023-03-17 00:15:48,451] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:15:48,451] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:15:48,451] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,451] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 0: [2023-03-17 00:15:48,451] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 5: [2023-03-17 00:15:48,451] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 6: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:15:48,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 2: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 2: [2023-03-17 00:15:48,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 2: [2023-03-17 00:15:48,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:15:48,450] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:15:48,450] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 2: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:15:48,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 2: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,451] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:15:48,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 2: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 7: [2023-03-17 00:15:48,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 7: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 7: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 3: [2023-03-17 00:15:48,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 1: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:15:48,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 3: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 4: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 4: [2023-03-17 00:15:48,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 1: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:15:48,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 4: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:15:48,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:15:48,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 1: [2023-03-17 00:15:48,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 1: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:15:48,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 0: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:15:48,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 5: [2023-03-17 00:15:48,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 4: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:15:48,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 1: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:15:48,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:15:48,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 1: [2023-03-17 00:15:48,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 4: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:15:48,455] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,455] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 4: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 5: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 5: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:15:48,455] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 3: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,455] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 5: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:15:48,455] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 4: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:15:48,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:15:48,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 1: [2023-03-17 00:15:48,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 4: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 1: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 3: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 2: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 3: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:15:48,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 4: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:15:48,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 00:15:48,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 4: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 4: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:15:48,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 0: [2023-03-17 00:15:48,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 4: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 0: [2023-03-17 00:15:48,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 1: [2023-03-17 00:15:48,458] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:15:48,458] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:15:48,458] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 6: [2023-03-17 00:15:48,458] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,459] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,459] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,459] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:15:48,459] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:15:48,459] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 00:15:48,459] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:15:48,459] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 7: [2023-03-17 00:15:48,459] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 7: [2023-03-17 00:15:48,459] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 2: [2023-03-17 00:15:48,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,458] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:15:48,458] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 6: [2023-03-17 00:15:48,458] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 2: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:15:48,458] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,458] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 2: [2023-03-17 00:15:48,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 00:15:48,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 6: [2023-03-17 00:15:48,458] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 2: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 2: [2023-03-17 00:15:48,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 6: [2023-03-17 00:15:48,458] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 6: [2023-03-17 00:15:48,458] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 0: [2023-03-17 00:15:48,473] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step4000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:15:48,473] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step4000 is ready now! 0: successfully saved checkpoint at iteration 4000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 403.75 7: iteration 4010/ 21553 | consumed samples: 1026560 | consumed tokens: 2102394880 | elapsed time per iteration (s): 0.35 | learning rate: 1.863E-04 | global batch size: 256 | lm loss: 3.901583E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 737.982 | TFLOPs: 25.83 | 7: iteration 4020/ 21553 | consumed samples: 1029120 | consumed tokens: 2107637760 | elapsed time per iteration (s): 0.30 | learning rate: 1.862E-04 | global batch size: 256 | lm loss: 3.888150E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.985 | TFLOPs: 30.35 | 7: iteration 4030/ 21553 | consumed samples: 1031680 | consumed tokens: 2112880640 | elapsed time per iteration (s): 0.30 | learning rate: 1.862E-04 | global batch size: 256 | lm loss: 3.890263E+00 | grad norm: 0.449 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.615 | TFLOPs: 30.30 | 7: iteration 4040/ 21553 | consumed samples: 1034240 | consumed tokens: 2118123520 | elapsed time per iteration (s): 0.30 | learning rate: 1.861E-04 | global batch size: 256 | lm loss: 3.889976E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.908 | TFLOPs: 30.31 | 7: iteration 4050/ 21553 | consumed samples: 1036800 | consumed tokens: 2123366400 | elapsed time per iteration (s): 0.30 | learning rate: 1.860E-04 | global batch size: 256 | lm loss: 3.889539E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.498 | TFLOPs: 30.33 | 7: iteration 4060/ 21553 | consumed samples: 1039360 | consumed tokens: 2128609280 | elapsed time per iteration (s): 0.30 | learning rate: 1.860E-04 | global batch size: 256 | lm loss: 3.890609E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.467 | TFLOPs: 30.33 | 7: iteration 4070/ 21553 | consumed samples: 1041920 | consumed tokens: 2133852160 | elapsed time per iteration (s): 0.30 | learning rate: 1.859E-04 | global batch size: 256 | lm loss: 3.887365E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.824 | TFLOPs: 30.35 | 7: iteration 4080/ 21553 | consumed samples: 1044480 | consumed tokens: 2139095040 | elapsed time per iteration (s): 0.30 | learning rate: 1.858E-04 | global batch size: 256 | lm loss: 3.887782E+00 | grad norm: 0.500 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.781 | TFLOPs: 30.34 | 7: iteration 4090/ 21553 | consumed samples: 1047040 | consumed tokens: 2144337920 | elapsed time per iteration (s): 0.30 | learning rate: 1.857E-04 | global batch size: 256 | lm loss: 3.881707E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.521 | TFLOPs: 30.33 | 7: iteration 4100/ 21553 | consumed samples: 1049600 | consumed tokens: 2149580800 | elapsed time per iteration (s): 0.30 | learning rate: 1.857E-04 | global batch size: 256 | lm loss: 3.885943E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.361 | TFLOPs: 30.33 | 7: iteration 4110/ 21553 | consumed samples: 1052160 | consumed tokens: 2154823680 | elapsed time per iteration (s): 0.30 | learning rate: 1.856E-04 | global batch size: 256 | lm loss: 3.881662E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.787 | TFLOPs: 30.34 | 7: iteration 4120/ 21553 | consumed samples: 1054720 | consumed tokens: 2160066560 | elapsed time per iteration (s): 0.30 | learning rate: 1.855E-04 | global batch size: 256 | lm loss: 3.880991E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.652 | TFLOPs: 30.27 | 7: iteration 4130/ 21553 | consumed samples: 1057280 | consumed tokens: 2165309440 | elapsed time per iteration (s): 0.30 | learning rate: 1.855E-04 | global batch size: 256 | lm loss: 3.879025E+00 | grad norm: 0.369 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.246 | TFLOPs: 30.32 | 7: iteration 4140/ 21553 | consumed samples: 1059840 | consumed tokens: 2170552320 | elapsed time per iteration (s): 0.30 | learning rate: 1.854E-04 | global batch size: 256 | lm loss: 3.870849E+00 | grad norm: 0.463 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.399 | TFLOPs: 30.33 | 7: iteration 4150/ 21553 | consumed samples: 1062400 | consumed tokens: 2175795200 | elapsed time per iteration (s): 0.30 | learning rate: 1.853E-04 | global batch size: 256 | lm loss: 3.883886E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.129 | TFLOPs: 30.32 | 7: iteration 4160/ 21553 | consumed samples: 1064960 | consumed tokens: 2181038080 | elapsed time per iteration (s): 0.30 | learning rate: 1.852E-04 | global batch size: 256 | lm loss: 3.874407E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.957 | TFLOPs: 30.31 | 7: iteration 4170/ 21553 | consumed samples: 1067520 | consumed tokens: 2186280960 | elapsed time per iteration (s): 0.30 | learning rate: 1.852E-04 | global batch size: 256 | lm loss: 3.876000E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.331 | TFLOPs: 30.33 | 7: iteration 4180/ 21553 | consumed samples: 1070080 | consumed tokens: 2191523840 | elapsed time per iteration (s): 0.30 | learning rate: 1.851E-04 | global batch size: 256 | lm loss: 3.872258E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.278 | TFLOPs: 30.33 | 7: iteration 4190/ 21553 | consumed samples: 1072640 | consumed tokens: 2196766720 | elapsed time per iteration (s): 0.30 | learning rate: 1.850E-04 | global batch size: 256 | lm loss: 3.870761E+00 | grad norm: 0.454 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.633 | TFLOPs: 30.34 | 7: iteration 4200/ 21553 | consumed samples: 1075200 | consumed tokens: 2202009600 | elapsed time per iteration (s): 0.30 | learning rate: 1.850E-04 | global batch size: 256 | lm loss: 3.879961E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.159 | TFLOPs: 30.32 | 7: iteration 4210/ 21553 | consumed samples: 1077760 | consumed tokens: 2207252480 | elapsed time per iteration (s): 0.30 | learning rate: 1.849E-04 | global batch size: 256 | lm loss: 3.877699E+00 | grad norm: 0.366 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.125 | TFLOPs: 30.32 | 7: iteration 4220/ 21553 | consumed samples: 1080320 | consumed tokens: 2212495360 | elapsed time per iteration (s): 0.30 | learning rate: 1.848E-04 | global batch size: 256 | lm loss: 3.865736E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.652 | TFLOPs: 30.34 | 7: iteration 4230/ 21553 | consumed samples: 1082880 | consumed tokens: 2217738240 | elapsed time per iteration (s): 0.30 | learning rate: 1.847E-04 | global batch size: 256 | lm loss: 3.871817E+00 | grad norm: 0.486 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.310 | TFLOPs: 30.33 | 7: iteration 4240/ 21553 | consumed samples: 1085440 | consumed tokens: 2222981120 | elapsed time per iteration (s): 0.30 | learning rate: 1.847E-04 | global batch size: 256 | lm loss: 3.870836E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.968 | TFLOPs: 30.32 | 7: iteration 4250/ 21553 | consumed samples: 1088000 | consumed tokens: 2228224000 | elapsed time per iteration (s): 0.30 | learning rate: 1.846E-04 | global batch size: 256 | lm loss: 3.864803E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.392 | TFLOPs: 30.33 | 7: iteration 4260/ 21553 | consumed samples: 1090560 | consumed tokens: 2233466880 | elapsed time per iteration (s): 0.30 | learning rate: 1.845E-04 | global batch size: 256 | lm loss: 3.864407E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.908 | TFLOPs: 30.31 | 7: iteration 4270/ 21553 | consumed samples: 1093120 | consumed tokens: 2238709760 | elapsed time per iteration (s): 0.30 | learning rate: 1.844E-04 | global batch size: 256 | lm loss: 3.864452E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.237 | TFLOPs: 30.32 | 7: iteration 4280/ 21553 | consumed samples: 1095680 | consumed tokens: 2243952640 | elapsed time per iteration (s): 0.30 | learning rate: 1.844E-04 | global batch size: 256 | lm loss: 3.862389E+00 | grad norm: 0.455 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.446 | TFLOPs: 30.33 | 7: iteration 4290/ 21553 | consumed samples: 1098240 | consumed tokens: 2249195520 | elapsed time per iteration (s): 0.30 | learning rate: 1.843E-04 | global batch size: 256 | lm loss: 3.859134E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.494 | TFLOPs: 30.33 | 7: iteration 4300/ 21553 | consumed samples: 1100800 | consumed tokens: 2254438400 | elapsed time per iteration (s): 0.30 | learning rate: 1.842E-04 | global batch size: 256 | lm loss: 3.857515E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.975 | TFLOPs: 30.32 | 7: iteration 4310/ 21553 | consumed samples: 1103360 | consumed tokens: 2259681280 | elapsed time per iteration (s): 0.30 | learning rate: 1.841E-04 | global batch size: 256 | lm loss: 3.856357E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.676 | TFLOPs: 30.30 | 7: iteration 4320/ 21553 | consumed samples: 1105920 | consumed tokens: 2264924160 | elapsed time per iteration (s): 0.30 | learning rate: 1.841E-04 | global batch size: 256 | lm loss: 3.862357E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.849 | TFLOPs: 30.35 | 7: iteration 4330/ 21553 | consumed samples: 1108480 | consumed tokens: 2270167040 | elapsed time per iteration (s): 0.30 | learning rate: 1.840E-04 | global batch size: 256 | lm loss: 3.850914E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.494 | TFLOPs: 30.26 | 7: iteration 4340/ 21553 | consumed samples: 1111040 | consumed tokens: 2275409920 | elapsed time per iteration (s): 0.30 | learning rate: 1.839E-04 | global batch size: 256 | lm loss: 3.858083E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.805 | TFLOPs: 30.03 | 7: iteration 4350/ 21553 | consumed samples: 1113600 | consumed tokens: 2280652800 | elapsed time per iteration (s): 0.30 | learning rate: 1.838E-04 | global batch size: 256 | lm loss: 3.851543E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.459 | TFLOPs: 30.33 | 7: iteration 4360/ 21553 | consumed samples: 1116160 | consumed tokens: 2285895680 | elapsed time per iteration (s): 0.30 | learning rate: 1.838E-04 | global batch size: 256 | lm loss: 3.863734E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.439 | TFLOPs: 30.33 | 7: iteration 4370/ 21553 | consumed samples: 1118720 | consumed tokens: 2291138560 | elapsed time per iteration (s): 0.30 | learning rate: 1.837E-04 | global batch size: 256 | lm loss: 3.857204E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.057 | TFLOPs: 30.32 | 7: iteration 4380/ 21553 | consumed samples: 1121280 | consumed tokens: 2296381440 | elapsed time per iteration (s): 0.30 | learning rate: 1.836E-04 | global batch size: 256 | lm loss: 3.857420E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 854.474 | TFLOPs: 29.91 | 7: iteration 4390/ 21553 | consumed samples: 1123840 | consumed tokens: 2301624320 | elapsed time per iteration (s): 0.30 | learning rate: 1.835E-04 | global batch size: 256 | lm loss: 3.850152E+00 | grad norm: 0.455 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.335 | TFLOPs: 30.33 | 7: iteration 4400/ 21553 | consumed samples: 1126400 | consumed tokens: 2306867200 | elapsed time per iteration (s): 0.30 | learning rate: 1.835E-04 | global batch size: 256 | lm loss: 3.858041E+00 | grad norm: 0.483 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.970 | TFLOPs: 30.35 | 7: iteration 4410/ 21553 | consumed samples: 1128960 | consumed tokens: 2312110080 | elapsed time per iteration (s): 0.30 | learning rate: 1.834E-04 | global batch size: 256 | lm loss: 3.846701E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.670 | TFLOPs: 30.34 | 7: iteration 4420/ 21553 | consumed samples: 1131520 | consumed tokens: 2317352960 | elapsed time per iteration (s): 0.30 | learning rate: 1.833E-04 | global batch size: 256 | lm loss: 3.846668E+00 | grad norm: 0.561 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.257 | TFLOPs: 30.33 | 7: iteration 4430/ 21553 | consumed samples: 1134080 | consumed tokens: 2322595840 | elapsed time per iteration (s): 0.30 | learning rate: 1.832E-04 | global batch size: 256 | lm loss: 3.852081E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.844 | TFLOPs: 30.31 | 7: iteration 4440/ 21553 | consumed samples: 1136640 | consumed tokens: 2327838720 | elapsed time per iteration (s): 0.30 | learning rate: 1.831E-04 | global batch size: 256 | lm loss: 3.852805E+00 | grad norm: 0.456 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.872 | TFLOPs: 30.31 | 7: iteration 4450/ 21553 | consumed samples: 1139200 | consumed tokens: 2333081600 | elapsed time per iteration (s): 0.30 | learning rate: 1.831E-04 | global batch size: 256 | lm loss: 3.850698E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.992 | TFLOPs: 30.32 | 7: iteration 4460/ 21553 | consumed samples: 1141760 | consumed tokens: 2338324480 | elapsed time per iteration (s): 0.30 | learning rate: 1.830E-04 | global batch size: 256 | lm loss: 3.844988E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.725 | TFLOPs: 30.31 | 7: iteration 4470/ 21553 | consumed samples: 1144320 | consumed tokens: 2343567360 | elapsed time per iteration (s): 0.30 | learning rate: 1.829E-04 | global batch size: 256 | lm loss: 3.843677E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.317 | TFLOPs: 30.29 | 7: iteration 4480/ 21553 | consumed samples: 1146880 | consumed tokens: 2348810240 | elapsed time per iteration (s): 0.30 | learning rate: 1.828E-04 | global batch size: 256 | lm loss: 3.835585E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.187 | TFLOPs: 30.32 | 7: iteration 4490/ 21553 | consumed samples: 1149440 | consumed tokens: 2354053120 | elapsed time per iteration (s): 0.30 | learning rate: 1.828E-04 | global batch size: 256 | lm loss: 3.850010E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.251 | TFLOPs: 30.33 | 7: iteration 4500/ 21553 | consumed samples: 1152000 | consumed tokens: 2359296000 | elapsed time per iteration (s): 0.30 | learning rate: 1.827E-04 | global batch size: 256 | lm loss: 3.851596E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.381 | TFLOPs: 30.15 | 7: iteration 4510/ 21553 | consumed samples: 1154560 | consumed tokens: 2364538880 | elapsed time per iteration (s): 0.30 | learning rate: 1.826E-04 | global batch size: 256 | lm loss: 3.835096E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.415 | TFLOPs: 30.33 | 7: iteration 4520/ 21553 | consumed samples: 1157120 | consumed tokens: 2369781760 | elapsed time per iteration (s): 0.30 | learning rate: 1.825E-04 | global batch size: 256 | lm loss: 3.833858E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.653 | TFLOPs: 30.34 | 7: iteration 4530/ 21553 | consumed samples: 1159680 | consumed tokens: 2375024640 | elapsed time per iteration (s): 0.30 | learning rate: 1.824E-04 | global batch size: 256 | lm loss: 3.839807E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.023 | TFLOPs: 30.32 | 7: iteration 4540/ 21553 | consumed samples: 1162240 | consumed tokens: 2380267520 | elapsed time per iteration (s): 0.30 | learning rate: 1.824E-04 | global batch size: 256 | lm loss: 3.836320E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.543 | TFLOPs: 30.34 | 7: iteration 4550/ 21553 | consumed samples: 1164800 | consumed tokens: 2385510400 | elapsed time per iteration (s): 0.30 | learning rate: 1.823E-04 | global batch size: 256 | lm loss: 3.826746E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.872 | TFLOPs: 30.35 | 7: iteration 4560/ 21553 | consumed samples: 1167360 | consumed tokens: 2390753280 | elapsed time per iteration (s): 0.30 | learning rate: 1.822E-04 | global batch size: 256 | lm loss: 3.835641E+00 | grad norm: 0.459 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.778 | TFLOPs: 30.34 | 7: iteration 4570/ 21553 | consumed samples: 1169920 | consumed tokens: 2395996160 | elapsed time per iteration (s): 0.30 | learning rate: 1.821E-04 | global batch size: 256 | lm loss: 3.841601E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.805 | TFLOPs: 30.34 | 7: iteration 4580/ 21553 | consumed samples: 1172480 | consumed tokens: 2401239040 | elapsed time per iteration (s): 0.30 | learning rate: 1.820E-04 | global batch size: 256 | lm loss: 3.832129E+00 | grad norm: 0.459 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.689 | TFLOPs: 30.34 | 7: iteration 4590/ 21553 | consumed samples: 1175040 | consumed tokens: 2406481920 | elapsed time per iteration (s): 0.30 | learning rate: 1.820E-04 | global batch size: 256 | lm loss: 3.834038E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.759 | TFLOPs: 30.34 | 7: iteration 4600/ 21553 | consumed samples: 1177600 | consumed tokens: 2411724800 | elapsed time per iteration (s): 0.30 | learning rate: 1.819E-04 | global batch size: 256 | lm loss: 3.833482E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.378 | TFLOPs: 30.33 | 7: iteration 4610/ 21553 | consumed samples: 1180160 | consumed tokens: 2416967680 | elapsed time per iteration (s): 0.30 | learning rate: 1.818E-04 | global batch size: 256 | lm loss: 3.831341E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.063 | TFLOPs: 30.32 | 7: iteration 4620/ 21553 | consumed samples: 1182720 | consumed tokens: 2422210560 | elapsed time per iteration (s): 0.30 | learning rate: 1.817E-04 | global batch size: 256 | lm loss: 3.833494E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.355 | TFLOPs: 30.33 | 7: iteration 4630/ 21553 | consumed samples: 1185280 | consumed tokens: 2427453440 | elapsed time per iteration (s): 0.30 | learning rate: 1.816E-04 | global batch size: 256 | lm loss: 3.830185E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.589 | TFLOPs: 30.34 | 7: iteration 4640/ 21553 | consumed samples: 1187840 | consumed tokens: 2432696320 | elapsed time per iteration (s): 0.30 | learning rate: 1.816E-04 | global batch size: 256 | lm loss: 3.825264E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.469 | TFLOPs: 30.33 | 7: iteration 4650/ 21553 | consumed samples: 1190400 | consumed tokens: 2437939200 | elapsed time per iteration (s): 0.30 | learning rate: 1.815E-04 | global batch size: 256 | lm loss: 3.821906E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.837 | TFLOPs: 30.35 | 7: iteration 4660/ 21553 | consumed samples: 1192960 | consumed tokens: 2443182080 | elapsed time per iteration (s): 0.30 | learning rate: 1.814E-04 | global batch size: 256 | lm loss: 3.828451E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.844 | TFLOPs: 30.31 | 7: iteration 4670/ 21553 | consumed samples: 1195520 | consumed tokens: 2448424960 | elapsed time per iteration (s): 0.30 | learning rate: 1.813E-04 | global batch size: 256 | lm loss: 3.821082E+00 | grad norm: 0.458 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.554 | TFLOPs: 30.34 | 7: iteration 4680/ 21553 | consumed samples: 1198080 | consumed tokens: 2453667840 | elapsed time per iteration (s): 0.30 | learning rate: 1.812E-04 | global batch size: 256 | lm loss: 3.820124E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.264 | TFLOPs: 30.33 | 7: iteration 4690/ 21553 | consumed samples: 1200640 | consumed tokens: 2458910720 | elapsed time per iteration (s): 0.30 | learning rate: 1.812E-04 | global batch size: 256 | lm loss: 3.824729E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.820 | TFLOPs: 30.31 | 7: iteration 4700/ 21553 | consumed samples: 1203200 | consumed tokens: 2464153600 | elapsed time per iteration (s): 0.30 | learning rate: 1.811E-04 | global batch size: 256 | lm loss: 3.827419E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.601 | TFLOPs: 30.30 | 7: iteration 4710/ 21553 | consumed samples: 1205760 | consumed tokens: 2469396480 | elapsed time per iteration (s): 0.30 | learning rate: 1.810E-04 | global batch size: 256 | lm loss: 3.830925E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.377 | TFLOPs: 30.29 | 7: iteration 4720/ 21553 | consumed samples: 1208320 | consumed tokens: 2474639360 | elapsed time per iteration (s): 0.30 | learning rate: 1.809E-04 | global batch size: 256 | lm loss: 3.819026E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 848.527 | TFLOPs: 29.70 | 7: iteration 4730/ 21553 | consumed samples: 1210880 | consumed tokens: 2479882240 | elapsed time per iteration (s): 0.30 | learning rate: 1.808E-04 | global batch size: 256 | lm loss: 3.816635E+00 | grad norm: 0.378 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.969 | TFLOPs: 30.00 | 7: iteration 4740/ 21553 | consumed samples: 1213440 | consumed tokens: 2485125120 | elapsed time per iteration (s): 0.30 | learning rate: 1.808E-04 | global batch size: 256 | lm loss: 3.821772E+00 | grad norm: 0.621 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.292 | TFLOPs: 30.22 | 7: iteration 4750/ 21553 | consumed samples: 1216000 | consumed tokens: 2490368000 | elapsed time per iteration (s): 0.30 | learning rate: 1.807E-04 | global batch size: 256 | lm loss: 3.819783E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.323 | TFLOPs: 30.33 | 7: iteration 4760/ 21553 | consumed samples: 1218560 | consumed tokens: 2495610880 | elapsed time per iteration (s): 0.30 | learning rate: 1.806E-04 | global batch size: 256 | lm loss: 3.821777E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.579 | TFLOPs: 30.30 | 7: iteration 4770/ 21553 | consumed samples: 1221120 | consumed tokens: 2500853760 | elapsed time per iteration (s): 0.30 | learning rate: 1.805E-04 | global batch size: 256 | lm loss: 3.815376E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.818 | TFLOPs: 30.31 | 7: iteration 4780/ 21553 | consumed samples: 1223680 | consumed tokens: 2506096640 | elapsed time per iteration (s): 0.30 | learning rate: 1.804E-04 | global batch size: 256 | lm loss: 3.814875E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.095 | TFLOPs: 30.32 | 7: iteration 4790/ 21553 | consumed samples: 1226240 | consumed tokens: 2511339520 | elapsed time per iteration (s): 0.30 | learning rate: 1.803E-04 | global batch size: 256 | lm loss: 3.816024E+00 | grad norm: 0.499 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.921 | TFLOPs: 30.28 | 7: iteration 4800/ 21553 | consumed samples: 1228800 | consumed tokens: 2516582400 | elapsed time per iteration (s): 0.30 | learning rate: 1.803E-04 | global batch size: 256 | lm loss: 3.807780E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.940 | TFLOPs: 30.28 | 7: iteration 4810/ 21553 | consumed samples: 1231360 | consumed tokens: 2521825280 | elapsed time per iteration (s): 0.30 | learning rate: 1.802E-04 | global batch size: 256 | lm loss: 3.814806E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.984 | TFLOPs: 30.28 | 7: iteration 4820/ 21553 | consumed samples: 1233920 | consumed tokens: 2527068160 | elapsed time per iteration (s): 0.30 | learning rate: 1.801E-04 | global batch size: 256 | lm loss: 3.805746E+00 | grad norm: 0.393 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.938 | TFLOPs: 30.28 | 7: iteration 4830/ 21553 | consumed samples: 1236480 | consumed tokens: 2532311040 | elapsed time per iteration (s): 0.30 | learning rate: 1.800E-04 | global batch size: 256 | lm loss: 3.810624E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.662 | TFLOPs: 30.27 | 7: iteration 4840/ 21553 | consumed samples: 1239040 | consumed tokens: 2537553920 | elapsed time per iteration (s): 0.30 | learning rate: 1.799E-04 | global batch size: 256 | lm loss: 3.809691E+00 | grad norm: 0.475 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.046 | TFLOPs: 30.28 | 7: iteration 4850/ 21553 | consumed samples: 1241600 | consumed tokens: 2542796800 | elapsed time per iteration (s): 0.30 | learning rate: 1.798E-04 | global batch size: 256 | lm loss: 3.810958E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.335 | TFLOPs: 30.29 | 7: iteration 4860/ 21553 | consumed samples: 1244160 | consumed tokens: 2548039680 | elapsed time per iteration (s): 0.30 | learning rate: 1.798E-04 | global batch size: 256 | lm loss: 3.798123E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.177 | TFLOPs: 30.29 | 7: iteration 4870/ 21553 | consumed samples: 1246720 | consumed tokens: 2553282560 | elapsed time per iteration (s): 0.30 | learning rate: 1.797E-04 | global batch size: 256 | lm loss: 3.809296E+00 | grad norm: 0.379 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.759 | TFLOPs: 30.27 | 7: iteration 4880/ 21553 | consumed samples: 1249280 | consumed tokens: 2558525440 | elapsed time per iteration (s): 0.30 | learning rate: 1.796E-04 | global batch size: 256 | lm loss: 3.802139E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.933 | TFLOPs: 30.28 | 7: iteration 4890/ 21553 | consumed samples: 1251840 | consumed tokens: 2563768320 | elapsed time per iteration (s): 0.30 | learning rate: 1.795E-04 | global batch size: 256 | lm loss: 3.806418E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.715 | TFLOPs: 30.27 | 7: iteration 4900/ 21553 | consumed samples: 1254400 | consumed tokens: 2569011200 | elapsed time per iteration (s): 0.30 | learning rate: 1.794E-04 | global batch size: 256 | lm loss: 3.807888E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.992 | TFLOPs: 30.28 | 7: iteration 4910/ 21553 | consumed samples: 1256960 | consumed tokens: 2574254080 | elapsed time per iteration (s): 0.30 | learning rate: 1.793E-04 | global batch size: 256 | lm loss: 3.803770E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.817 | TFLOPs: 30.27 | 7: iteration 4920/ 21553 | consumed samples: 1259520 | consumed tokens: 2579496960 | elapsed time per iteration (s): 0.30 | learning rate: 1.793E-04 | global batch size: 256 | lm loss: 3.805851E+00 | grad norm: 0.380 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.030 | TFLOPs: 30.28 | 7: iteration 4930/ 21553 | consumed samples: 1262080 | consumed tokens: 2584739840 | elapsed time per iteration (s): 0.30 | learning rate: 1.792E-04 | global batch size: 256 | lm loss: 3.799316E+00 | grad norm: 0.489 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.031 | TFLOPs: 30.28 | 7: iteration 4940/ 21553 | consumed samples: 1264640 | consumed tokens: 2589982720 | elapsed time per iteration (s): 0.30 | learning rate: 1.791E-04 | global batch size: 256 | lm loss: 3.799989E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.485 | TFLOPs: 30.30 | 7: iteration 4950/ 21553 | consumed samples: 1267200 | consumed tokens: 2595225600 | elapsed time per iteration (s): 0.30 | learning rate: 1.790E-04 | global batch size: 256 | lm loss: 3.798538E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.818 | TFLOPs: 30.31 | 7: iteration 4960/ 21553 | consumed samples: 1269760 | consumed tokens: 2600468480 | elapsed time per iteration (s): 0.30 | learning rate: 1.789E-04 | global batch size: 256 | lm loss: 3.794316E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.831 | TFLOPs: 30.31 | 7: iteration 4970/ 21553 | consumed samples: 1272320 | consumed tokens: 2605711360 | elapsed time per iteration (s): 0.30 | learning rate: 1.788E-04 | global batch size: 256 | lm loss: 3.796478E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.074 | TFLOPs: 30.32 | 7: iteration 4980/ 21553 | consumed samples: 1274880 | consumed tokens: 2610954240 | elapsed time per iteration (s): 0.30 | learning rate: 1.787E-04 | global batch size: 256 | lm loss: 3.799132E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.589 | TFLOPs: 30.30 | 7: iteration 4990/ 21553 | consumed samples: 1277440 | consumed tokens: 2616197120 | elapsed time per iteration (s): 0.30 | learning rate: 1.787E-04 | global batch size: 256 | lm loss: 3.799219E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.059 | TFLOPs: 30.32 | 7: iteration 5000/ 21553 | consumed samples: 1280000 | consumed tokens: 2621440000 | elapsed time per iteration (s): 0.30 | learning rate: 1.786E-04 | global batch size: 256 | lm loss: 3.799769E+00 | grad norm: 0.472 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.335 | TFLOPs: 30.29 | 7: ----------------------------------------------------------------------------------------------- 7: validation loss at iteration 5000 | lm loss value: 3.875289E+00 | lm loss PPL: 4.819662E+01 | 7: ----------------------------------------------------------------------------------------------- 0: saving checkpoint at iteration 5000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:20:44,371] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step5000 is begin to save! 0: [2023-03-17 00:20:44,374] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:20:44,465] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:20:44,466] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:20:44,483] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:20:44,483] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:20:44,498] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:20:44,498] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:20:44,513] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:20:44,513] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:20:44,528] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:20:44,528] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:20:44,543] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:20:44,543] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:20:44,558] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:20:44,558] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:20:44,573] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:20:44,573] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:20:44,588] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:20:44,588] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:20:44,603] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:20:44,603] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:20:44,618] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:20:44,619] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:20:44,633] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:20:44,634] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:20:44,649] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:20:44,649] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:20:44,664] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:20:44,664] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:20:44,679] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:20:44,679] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:20:44,694] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:20:44,694] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:20:44,695] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:20:44,696] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step5000/mp_rank_00_model_states.pt 0: [2023-03-17 00:20:44,696] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:20:44,698] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:20:44,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:20:44,746] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:20:44,750] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:20:44,750] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,750] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 2: [2023-03-17 00:20:44,750] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,750] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 00:20:44,750] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 3: [2023-03-17 00:20:44,751] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:20:44,751] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,751] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 0: [2023-03-17 00:20:44,751] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:20:44,751] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:20:44,751] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 1: [2023-03-17 00:20:44,751] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:20:44,752] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 1: [2023-03-17 00:20:44,751] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:20:44,752] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 0: [2023-03-17 00:20:44,751] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,751] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:20:44,752] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,752] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:20:44,752] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:20:44,753] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:20:44,753] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 2: [2023-03-17 00:20:44,753] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,753] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:20:44,753] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 3: [2023-03-17 00:20:44,754] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:20:44,754] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,754] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 2: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 3: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 5: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 1: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 5: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 1: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 1: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 1: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 5: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 2: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,756] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 3: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:20:44,756] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 6: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 6: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:20:44,756] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 6: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:20:44,756] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 6: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:20:44,756] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 6: [2023-03-17 00:20:44,756] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 7: [2023-03-17 00:20:44,752] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,752] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,752] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:20:44,752] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 7: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 4: [2023-03-17 00:20:44,754] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,754] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:20:44,754] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 7: [2023-03-17 00:20:44,755] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 4: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,756] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,755] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 7: [2023-03-17 00:20:44,756] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,756] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 1: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:20:44,757] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 2: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,757] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:20:44,757] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 2: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:20:44,757] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 0: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:20:44,757] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 00:20:44,757] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 0: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 1: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:20:44,757] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 5: [2023-03-17 00:20:44,758] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:20:44,758] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:20:44,758] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 5: [2023-03-17 00:20:44,758] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:20:44,758] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:20:44,758] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:20:44,758] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 00:20:44,758] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 5: [2023-03-17 00:20:44,758] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 5: [2023-03-17 00:20:44,758] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:20:44,758] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 00:20:44,758] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 5: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:20:44,759] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 2: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,760] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 0: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 0: [2023-03-17 00:20:44,760] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 6: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:20:44,760] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 1: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:20:44,760] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 0: [2023-03-17 00:20:44,761] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:20:44,761] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 4: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:20:44,759] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:20:44,761] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,759] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:20:44,759] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 4: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:20:44,760] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 0: [2023-03-17 00:20:44,761] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:20:44,761] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 00:20:44,761] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 3: [2023-03-17 00:20:44,761] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:20:44,761] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,761] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 2: [2023-03-17 00:20:44,762] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,762] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,762] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:20:44,762] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 3: [2023-03-17 00:20:44,762] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,762] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 6: [2023-03-17 00:20:44,763] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:20:44,763] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,763] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:20:44,763] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:20:44,763] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 3: [2023-03-17 00:20:44,763] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 6: [2023-03-17 00:20:44,763] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 3: [2023-03-17 00:20:44,763] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 6: [2023-03-17 00:20:44,763] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 7: [2023-03-17 00:20:44,757] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,757] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 7: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:20:44,759] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 7: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:20:44,759] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,759] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 7: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:20:44,760] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:20:44,760] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 0: [2023-03-17 00:20:44,769] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step5000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:20:44,769] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step5000 is ready now! 0: successfully saved checkpoint at iteration 5000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 411.10 7: iteration 5010/ 21553 | consumed samples: 1282560 | consumed tokens: 2626682880 | elapsed time per iteration (s): 0.35 | learning rate: 1.785E-04 | global batch size: 256 | lm loss: 3.795134E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 739.819 | TFLOPs: 25.90 | 7: iteration 5020/ 21553 | consumed samples: 1285120 | consumed tokens: 2631925760 | elapsed time per iteration (s): 0.30 | learning rate: 1.784E-04 | global batch size: 256 | lm loss: 3.791795E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.615 | TFLOPs: 30.30 | 7: iteration 5030/ 21553 | consumed samples: 1287680 | consumed tokens: 2637168640 | elapsed time per iteration (s): 0.30 | learning rate: 1.783E-04 | global batch size: 256 | lm loss: 3.795049E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.518 | TFLOPs: 30.30 | 7: iteration 5040/ 21553 | consumed samples: 1290240 | consumed tokens: 2642411520 | elapsed time per iteration (s): 0.30 | learning rate: 1.782E-04 | global batch size: 256 | lm loss: 3.792812E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.311 | TFLOPs: 30.29 | 7: iteration 5050/ 21553 | consumed samples: 1292800 | consumed tokens: 2647654400 | elapsed time per iteration (s): 0.30 | learning rate: 1.781E-04 | global batch size: 256 | lm loss: 3.784832E+00 | grad norm: 0.386 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.477 | TFLOPs: 30.30 | 7: iteration 5060/ 21553 | consumed samples: 1295360 | consumed tokens: 2652897280 | elapsed time per iteration (s): 0.30 | learning rate: 1.781E-04 | global batch size: 256 | lm loss: 3.792677E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.110 | TFLOPs: 30.32 | 7: iteration 5070/ 21553 | consumed samples: 1297920 | consumed tokens: 2658140160 | elapsed time per iteration (s): 0.30 | learning rate: 1.780E-04 | global batch size: 256 | lm loss: 3.789527E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.154 | TFLOPs: 30.32 | 7: iteration 5080/ 21553 | consumed samples: 1300480 | consumed tokens: 2663383040 | elapsed time per iteration (s): 0.30 | learning rate: 1.779E-04 | global batch size: 256 | lm loss: 3.793259E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.572 | TFLOPs: 30.30 | 7: iteration 5090/ 21553 | consumed samples: 1303040 | consumed tokens: 2668625920 | elapsed time per iteration (s): 0.30 | learning rate: 1.778E-04 | global batch size: 256 | lm loss: 3.789966E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.077 | TFLOPs: 30.32 | 7: iteration 5100/ 21553 | consumed samples: 1305600 | consumed tokens: 2673868800 | elapsed time per iteration (s): 0.30 | learning rate: 1.777E-04 | global batch size: 256 | lm loss: 3.782458E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.101 | TFLOPs: 30.32 | 7: iteration 5110/ 21553 | consumed samples: 1308160 | consumed tokens: 2679111680 | elapsed time per iteration (s): 0.30 | learning rate: 1.776E-04 | global batch size: 256 | lm loss: 3.780538E+00 | grad norm: 0.393 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.467 | TFLOPs: 30.30 | 7: iteration 5120/ 21553 | consumed samples: 1310720 | consumed tokens: 2684354560 | elapsed time per iteration (s): 0.30 | learning rate: 1.775E-04 | global batch size: 256 | lm loss: 3.783534E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.260 | TFLOPs: 30.15 | 7: iteration 5130/ 21553 | consumed samples: 1313280 | consumed tokens: 2689597440 | elapsed time per iteration (s): 0.30 | learning rate: 1.775E-04 | global batch size: 256 | lm loss: 3.783564E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.602 | TFLOPs: 30.30 | 7: iteration 5140/ 21553 | consumed samples: 1315840 | consumed tokens: 2694840320 | elapsed time per iteration (s): 0.30 | learning rate: 1.774E-04 | global batch size: 256 | lm loss: 3.786213E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.293 | TFLOPs: 30.29 | 7: iteration 5150/ 21553 | consumed samples: 1318400 | consumed tokens: 2700083200 | elapsed time per iteration (s): 0.30 | learning rate: 1.773E-04 | global batch size: 256 | lm loss: 3.774038E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.526 | TFLOPs: 30.30 | 7: iteration 5160/ 21553 | consumed samples: 1320960 | consumed tokens: 2705326080 | elapsed time per iteration (s): 0.30 | learning rate: 1.772E-04 | global batch size: 256 | lm loss: 3.775453E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.946 | TFLOPs: 30.31 | 7: iteration 5170/ 21553 | consumed samples: 1323520 | consumed tokens: 2710568960 | elapsed time per iteration (s): 0.30 | learning rate: 1.771E-04 | global batch size: 256 | lm loss: 3.779576E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.094 | TFLOPs: 30.32 | 7: iteration 5180/ 21553 | consumed samples: 1326080 | consumed tokens: 2715811840 | elapsed time per iteration (s): 0.30 | learning rate: 1.770E-04 | global batch size: 256 | lm loss: 3.773734E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.757 | TFLOPs: 30.31 | 7: iteration 5190/ 21553 | consumed samples: 1328640 | consumed tokens: 2721054720 | elapsed time per iteration (s): 0.30 | learning rate: 1.769E-04 | global batch size: 256 | lm loss: 3.778241E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.001 | TFLOPs: 30.32 | 7: iteration 5200/ 21553 | consumed samples: 1331200 | consumed tokens: 2726297600 | elapsed time per iteration (s): 0.30 | learning rate: 1.768E-04 | global batch size: 256 | lm loss: 3.776627E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.507 | TFLOPs: 30.05 | 7: iteration 5210/ 21553 | consumed samples: 1333760 | consumed tokens: 2731540480 | elapsed time per iteration (s): 0.30 | learning rate: 1.767E-04 | global batch size: 256 | lm loss: 3.778299E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.369 | TFLOPs: 30.29 | 7: iteration 5220/ 21553 | consumed samples: 1336320 | consumed tokens: 2736783360 | elapsed time per iteration (s): 0.30 | learning rate: 1.767E-04 | global batch size: 256 | lm loss: 3.777932E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.559 | TFLOPs: 30.30 | 7: iteration 5230/ 21553 | consumed samples: 1338880 | consumed tokens: 2742026240 | elapsed time per iteration (s): 0.30 | learning rate: 1.766E-04 | global batch size: 256 | lm loss: 3.771100E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.380 | TFLOPs: 30.29 | 7: iteration 5240/ 21553 | consumed samples: 1341440 | consumed tokens: 2747269120 | elapsed time per iteration (s): 0.30 | learning rate: 1.765E-04 | global batch size: 256 | lm loss: 3.770530E+00 | grad norm: 0.386 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.689 | TFLOPs: 30.06 | 7: iteration 5250/ 21553 | consumed samples: 1344000 | consumed tokens: 2752512000 | elapsed time per iteration (s): 0.30 | learning rate: 1.764E-04 | global batch size: 256 | lm loss: 3.772070E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.891 | TFLOPs: 30.31 | 7: iteration 5260/ 21553 | consumed samples: 1346560 | consumed tokens: 2757754880 | elapsed time per iteration (s): 0.30 | learning rate: 1.763E-04 | global batch size: 256 | lm loss: 3.776096E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.312 | TFLOPs: 30.29 | 7: iteration 5270/ 21553 | consumed samples: 1349120 | consumed tokens: 2762997760 | elapsed time per iteration (s): 0.30 | learning rate: 1.762E-04 | global batch size: 256 | lm loss: 3.768348E+00 | grad norm: 0.375 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.605 | TFLOPs: 30.30 | 7: iteration 5280/ 21553 | consumed samples: 1351680 | consumed tokens: 2768240640 | elapsed time per iteration (s): 0.30 | learning rate: 1.761E-04 | global batch size: 256 | lm loss: 3.771711E+00 | grad norm: 0.461 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.558 | TFLOPs: 30.30 | 7: iteration 5290/ 21553 | consumed samples: 1354240 | consumed tokens: 2773483520 | elapsed time per iteration (s): 0.30 | learning rate: 1.760E-04 | global batch size: 256 | lm loss: 3.774917E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.703 | TFLOPs: 30.31 | 7: iteration 5300/ 21553 | consumed samples: 1356800 | consumed tokens: 2778726400 | elapsed time per iteration (s): 0.30 | learning rate: 1.759E-04 | global batch size: 256 | lm loss: 3.768122E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.834 | TFLOPs: 30.31 | 7: iteration 5310/ 21553 | consumed samples: 1359360 | consumed tokens: 2783969280 | elapsed time per iteration (s): 0.30 | learning rate: 1.758E-04 | global batch size: 256 | lm loss: 3.768320E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.709 | TFLOPs: 30.31 | 7: iteration 5320/ 21553 | consumed samples: 1361920 | consumed tokens: 2789212160 | elapsed time per iteration (s): 0.30 | learning rate: 1.758E-04 | global batch size: 256 | lm loss: 3.761037E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.440 | TFLOPs: 30.33 | 7: iteration 5330/ 21553 | consumed samples: 1364480 | consumed tokens: 2794455040 | elapsed time per iteration (s): 0.30 | learning rate: 1.757E-04 | global batch size: 256 | lm loss: 3.764822E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.318 | TFLOPs: 30.33 | 7: iteration 5340/ 21553 | consumed samples: 1367040 | consumed tokens: 2799697920 | elapsed time per iteration (s): 0.30 | learning rate: 1.756E-04 | global batch size: 256 | lm loss: 3.765288E+00 | grad norm: 0.372 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.257 | TFLOPs: 30.33 | 7: iteration 5350/ 21553 | consumed samples: 1369600 | consumed tokens: 2804940800 | elapsed time per iteration (s): 0.30 | learning rate: 1.755E-04 | global batch size: 256 | lm loss: 3.768890E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.144 | TFLOPs: 30.32 | 7: iteration 5360/ 21553 | consumed samples: 1372160 | consumed tokens: 2810183680 | elapsed time per iteration (s): 0.30 | learning rate: 1.754E-04 | global batch size: 256 | lm loss: 3.766360E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.390 | TFLOPs: 30.33 | 7: iteration 5370/ 21553 | consumed samples: 1374720 | consumed tokens: 2815426560 | elapsed time per iteration (s): 0.30 | learning rate: 1.753E-04 | global batch size: 256 | lm loss: 3.769617E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.344 | TFLOPs: 30.33 | 7: iteration 5380/ 21553 | consumed samples: 1377280 | consumed tokens: 2820669440 | elapsed time per iteration (s): 0.30 | learning rate: 1.752E-04 | global batch size: 256 | lm loss: 3.760294E+00 | grad norm: 0.379 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.576 | TFLOPs: 30.34 | 7: iteration 5390/ 21553 | consumed samples: 1379840 | consumed tokens: 2825912320 | elapsed time per iteration (s): 0.30 | learning rate: 1.751E-04 | global batch size: 256 | lm loss: 3.768237E+00 | grad norm: 0.459 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.190 | TFLOPs: 30.32 | 7: iteration 5400/ 21553 | consumed samples: 1382400 | consumed tokens: 2831155200 | elapsed time per iteration (s): 0.30 | learning rate: 1.750E-04 | global batch size: 256 | lm loss: 3.758942E+00 | grad norm: 0.380 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.963 | TFLOPs: 30.31 | 7: iteration 5410/ 21553 | consumed samples: 1384960 | consumed tokens: 2836398080 | elapsed time per iteration (s): 0.30 | learning rate: 1.749E-04 | global batch size: 256 | lm loss: 3.763150E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.281 | TFLOPs: 30.33 | 7: iteration 5420/ 21553 | consumed samples: 1387520 | consumed tokens: 2841640960 | elapsed time per iteration (s): 0.30 | learning rate: 1.748E-04 | global batch size: 256 | lm loss: 3.763745E+00 | grad norm: 0.528 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.659 | TFLOPs: 30.34 | 7: iteration 5430/ 21553 | consumed samples: 1390080 | consumed tokens: 2846883840 | elapsed time per iteration (s): 0.30 | learning rate: 1.748E-04 | global batch size: 256 | lm loss: 3.767034E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.569 | TFLOPs: 30.34 | 7: iteration 5440/ 21553 | consumed samples: 1392640 | consumed tokens: 2852126720 | elapsed time per iteration (s): 0.30 | learning rate: 1.747E-04 | global batch size: 256 | lm loss: 3.763083E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.750 | TFLOPs: 30.31 | 7: iteration 5450/ 21553 | consumed samples: 1395200 | consumed tokens: 2857369600 | elapsed time per iteration (s): 0.30 | learning rate: 1.746E-04 | global batch size: 256 | lm loss: 3.749441E+00 | grad norm: 0.387 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.370 | TFLOPs: 30.33 | 7: iteration 5460/ 21553 | consumed samples: 1397760 | consumed tokens: 2862612480 | elapsed time per iteration (s): 0.30 | learning rate: 1.745E-04 | global batch size: 256 | lm loss: 3.759786E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.112 | TFLOPs: 30.32 | 7: iteration 5470/ 21553 | consumed samples: 1400320 | consumed tokens: 2867855360 | elapsed time per iteration (s): 0.30 | learning rate: 1.744E-04 | global batch size: 256 | lm loss: 3.748853E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.766 | TFLOPs: 30.31 | 7: iteration 5480/ 21553 | consumed samples: 1402880 | consumed tokens: 2873098240 | elapsed time per iteration (s): 0.30 | learning rate: 1.743E-04 | global batch size: 256 | lm loss: 3.751125E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.180 | TFLOPs: 30.32 | 7: iteration 5490/ 21553 | consumed samples: 1405440 | consumed tokens: 2878341120 | elapsed time per iteration (s): 0.30 | learning rate: 1.742E-04 | global batch size: 256 | lm loss: 3.753121E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.162 | TFLOPs: 30.29 | 7: iteration 5500/ 21553 | consumed samples: 1408000 | consumed tokens: 2883584000 | elapsed time per iteration (s): 0.30 | learning rate: 1.741E-04 | global batch size: 256 | lm loss: 3.751447E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.433 | TFLOPs: 30.30 | 7: iteration 5510/ 21553 | consumed samples: 1410560 | consumed tokens: 2888826880 | elapsed time per iteration (s): 0.30 | learning rate: 1.740E-04 | global batch size: 256 | lm loss: 3.753096E+00 | grad norm: 0.460 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.444 | TFLOPs: 30.30 | 7: iteration 5520/ 21553 | consumed samples: 1413120 | consumed tokens: 2894069760 | elapsed time per iteration (s): 0.30 | learning rate: 1.739E-04 | global batch size: 256 | lm loss: 3.752855E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.444 | TFLOPs: 30.30 | 7: iteration 5530/ 21553 | consumed samples: 1415680 | consumed tokens: 2899312640 | elapsed time per iteration (s): 0.30 | learning rate: 1.738E-04 | global batch size: 256 | lm loss: 3.752536E+00 | grad norm: 0.451 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.800 | TFLOPs: 30.27 | 7: iteration 5540/ 21553 | consumed samples: 1418240 | consumed tokens: 2904555520 | elapsed time per iteration (s): 0.30 | learning rate: 1.737E-04 | global batch size: 256 | lm loss: 3.751715E+00 | grad norm: 0.397 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.688 | TFLOPs: 30.31 | 7: iteration 5550/ 21553 | consumed samples: 1420800 | consumed tokens: 2909798400 | elapsed time per iteration (s): 0.30 | learning rate: 1.736E-04 | global batch size: 256 | lm loss: 3.755360E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.413 | TFLOPs: 30.30 | 7: iteration 5560/ 21553 | consumed samples: 1423360 | consumed tokens: 2915041280 | elapsed time per iteration (s): 0.30 | learning rate: 1.735E-04 | global batch size: 256 | lm loss: 3.749390E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.234 | TFLOPs: 30.29 | 7: iteration 5570/ 21553 | consumed samples: 1425920 | consumed tokens: 2920284160 | elapsed time per iteration (s): 0.30 | learning rate: 1.735E-04 | global batch size: 256 | lm loss: 3.757129E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.972 | TFLOPs: 30.32 | 7: iteration 5580/ 21553 | consumed samples: 1428480 | consumed tokens: 2925527040 | elapsed time per iteration (s): 0.30 | learning rate: 1.734E-04 | global batch size: 256 | lm loss: 3.745322E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.192 | TFLOPs: 30.29 | 7: iteration 5590/ 21553 | consumed samples: 1431040 | consumed tokens: 2930769920 | elapsed time per iteration (s): 0.30 | learning rate: 1.733E-04 | global batch size: 256 | lm loss: 3.749377E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.645 | TFLOPs: 30.27 | 7: iteration 5600/ 21553 | consumed samples: 1433600 | consumed tokens: 2936012800 | elapsed time per iteration (s): 0.30 | learning rate: 1.732E-04 | global batch size: 256 | lm loss: 3.750634E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 855.942 | TFLOPs: 29.96 | 7: iteration 5610/ 21553 | consumed samples: 1436160 | consumed tokens: 2941255680 | elapsed time per iteration (s): 0.30 | learning rate: 1.731E-04 | global batch size: 256 | lm loss: 3.743647E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.973 | TFLOPs: 30.28 | 7: iteration 5620/ 21553 | consumed samples: 1438720 | consumed tokens: 2946498560 | elapsed time per iteration (s): 0.30 | learning rate: 1.730E-04 | global batch size: 256 | lm loss: 3.743833E+00 | grad norm: 0.472 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.166 | TFLOPs: 30.29 | 7: iteration 5630/ 21553 | consumed samples: 1441280 | consumed tokens: 2951741440 | elapsed time per iteration (s): 0.30 | learning rate: 1.729E-04 | global batch size: 256 | lm loss: 3.745379E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.345 | TFLOPs: 30.26 | 7: iteration 5640/ 21553 | consumed samples: 1443840 | consumed tokens: 2956984320 | elapsed time per iteration (s): 0.30 | learning rate: 1.728E-04 | global batch size: 256 | lm loss: 3.744695E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.927 | TFLOPs: 30.28 | 7: iteration 5650/ 21553 | consumed samples: 1446400 | consumed tokens: 2962227200 | elapsed time per iteration (s): 0.30 | learning rate: 1.727E-04 | global batch size: 256 | lm loss: 3.743687E+00 | grad norm: 0.386 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.980 | TFLOPs: 30.32 | 7: iteration 5660/ 21553 | consumed samples: 1448960 | consumed tokens: 2967470080 | elapsed time per iteration (s): 0.30 | learning rate: 1.726E-04 | global batch size: 256 | lm loss: 3.739328E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.500 | TFLOPs: 30.05 | 7: iteration 5670/ 21553 | consumed samples: 1451520 | consumed tokens: 2972712960 | elapsed time per iteration (s): 0.30 | learning rate: 1.725E-04 | global batch size: 256 | lm loss: 3.732423E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.709 | TFLOPs: 30.31 | 7: iteration 5680/ 21553 | consumed samples: 1454080 | consumed tokens: 2977955840 | elapsed time per iteration (s): 0.30 | learning rate: 1.724E-04 | global batch size: 256 | lm loss: 3.741112E+00 | grad norm: 0.475 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.771 | TFLOPs: 30.31 | 7: iteration 5690/ 21553 | consumed samples: 1456640 | consumed tokens: 2983198720 | elapsed time per iteration (s): 0.30 | learning rate: 1.723E-04 | global batch size: 256 | lm loss: 3.739669E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.101 | TFLOPs: 30.32 | 7: iteration 5700/ 21553 | consumed samples: 1459200 | consumed tokens: 2988441600 | elapsed time per iteration (s): 0.30 | learning rate: 1.722E-04 | global batch size: 256 | lm loss: 3.737249E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.604 | TFLOPs: 30.30 | 7: iteration 5710/ 21553 | consumed samples: 1461760 | consumed tokens: 2993684480 | elapsed time per iteration (s): 0.30 | learning rate: 1.721E-04 | global batch size: 256 | lm loss: 3.744272E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.902 | TFLOPs: 30.31 | 7: iteration 5720/ 21553 | consumed samples: 1464320 | consumed tokens: 2998927360 | elapsed time per iteration (s): 0.30 | learning rate: 1.720E-04 | global batch size: 256 | lm loss: 3.740267E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.794 | TFLOPs: 30.31 | 7: iteration 5730/ 21553 | consumed samples: 1466880 | consumed tokens: 3004170240 | elapsed time per iteration (s): 0.30 | learning rate: 1.719E-04 | global batch size: 256 | lm loss: 3.746574E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.849 | TFLOPs: 30.31 | 7: iteration 5740/ 21553 | consumed samples: 1469440 | consumed tokens: 3009413120 | elapsed time per iteration (s): 0.30 | learning rate: 1.718E-04 | global batch size: 256 | lm loss: 3.741192E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.799 | TFLOPs: 30.31 | 7: iteration 5750/ 21553 | consumed samples: 1472000 | consumed tokens: 3014656000 | elapsed time per iteration (s): 0.30 | learning rate: 1.717E-04 | global batch size: 256 | lm loss: 3.736757E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.699 | TFLOPs: 30.31 | 7: iteration 5760/ 21553 | consumed samples: 1474560 | consumed tokens: 3019898880 | elapsed time per iteration (s): 0.30 | learning rate: 1.716E-04 | global batch size: 256 | lm loss: 3.737260E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.707 | TFLOPs: 30.31 | 7: iteration 5770/ 21553 | consumed samples: 1477120 | consumed tokens: 3025141760 | elapsed time per iteration (s): 0.30 | learning rate: 1.715E-04 | global batch size: 256 | lm loss: 3.738290E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.225 | TFLOPs: 30.29 | 7: iteration 5780/ 21553 | consumed samples: 1479680 | consumed tokens: 3030384640 | elapsed time per iteration (s): 0.30 | learning rate: 1.714E-04 | global batch size: 256 | lm loss: 3.735443E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.536 | TFLOPs: 30.30 | 7: iteration 5790/ 21553 | consumed samples: 1482240 | consumed tokens: 3035627520 | elapsed time per iteration (s): 0.30 | learning rate: 1.714E-04 | global batch size: 256 | lm loss: 3.735356E+00 | grad norm: 0.455 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.313 | TFLOPs: 30.29 | 7: iteration 5800/ 21553 | consumed samples: 1484800 | consumed tokens: 3040870400 | elapsed time per iteration (s): 0.30 | learning rate: 1.713E-04 | global batch size: 256 | lm loss: 3.725345E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.983 | TFLOPs: 30.28 | 7: iteration 5810/ 21553 | consumed samples: 1487360 | consumed tokens: 3046113280 | elapsed time per iteration (s): 0.30 | learning rate: 1.712E-04 | global batch size: 256 | lm loss: 3.728672E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.385 | TFLOPs: 30.29 | 7: iteration 5820/ 21553 | consumed samples: 1489920 | consumed tokens: 3051356160 | elapsed time per iteration (s): 0.30 | learning rate: 1.711E-04 | global batch size: 256 | lm loss: 3.724911E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.631 | TFLOPs: 30.30 | 7: iteration 5830/ 21553 | consumed samples: 1492480 | consumed tokens: 3056599040 | elapsed time per iteration (s): 0.30 | learning rate: 1.710E-04 | global batch size: 256 | lm loss: 3.727721E+00 | grad norm: 0.393 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.379 | TFLOPs: 30.29 | 7: iteration 5840/ 21553 | consumed samples: 1495040 | consumed tokens: 3061841920 | elapsed time per iteration (s): 0.30 | learning rate: 1.709E-04 | global batch size: 256 | lm loss: 3.732732E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.227 | TFLOPs: 30.29 | 7: iteration 5850/ 21553 | consumed samples: 1497600 | consumed tokens: 3067084800 | elapsed time per iteration (s): 0.30 | learning rate: 1.708E-04 | global batch size: 256 | lm loss: 3.732324E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.447 | TFLOPs: 30.30 | 7: iteration 5860/ 21553 | consumed samples: 1500160 | consumed tokens: 3072327680 | elapsed time per iteration (s): 0.30 | learning rate: 1.707E-04 | global batch size: 256 | lm loss: 3.734295E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.430 | TFLOPs: 30.30 | 7: iteration 5870/ 21553 | consumed samples: 1502720 | consumed tokens: 3077570560 | elapsed time per iteration (s): 0.30 | learning rate: 1.706E-04 | global batch size: 256 | lm loss: 3.721024E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.601 | TFLOPs: 30.30 | 7: iteration 5880/ 21553 | consumed samples: 1505280 | consumed tokens: 3082813440 | elapsed time per iteration (s): 0.30 | learning rate: 1.705E-04 | global batch size: 256 | lm loss: 3.719540E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.872 | TFLOPs: 30.31 | 7: iteration 5890/ 21553 | consumed samples: 1507840 | consumed tokens: 3088056320 | elapsed time per iteration (s): 0.30 | learning rate: 1.704E-04 | global batch size: 256 | lm loss: 3.728692E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.864 | TFLOPs: 30.31 | 7: iteration 5900/ 21553 | consumed samples: 1510400 | consumed tokens: 3093299200 | elapsed time per iteration (s): 0.30 | learning rate: 1.703E-04 | global batch size: 256 | lm loss: 3.730952E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.108 | TFLOPs: 30.32 | 7: iteration 5910/ 21553 | consumed samples: 1512960 | consumed tokens: 3098542080 | elapsed time per iteration (s): 0.30 | learning rate: 1.702E-04 | global batch size: 256 | lm loss: 3.726759E+00 | grad norm: 0.384 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.916 | TFLOPs: 30.31 | 7: iteration 5920/ 21553 | consumed samples: 1515520 | consumed tokens: 3103784960 | elapsed time per iteration (s): 0.30 | learning rate: 1.701E-04 | global batch size: 256 | lm loss: 3.727504E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.793 | TFLOPs: 30.31 | 7: iteration 5930/ 21553 | consumed samples: 1518080 | consumed tokens: 3109027840 | elapsed time per iteration (s): 0.30 | learning rate: 1.700E-04 | global batch size: 256 | lm loss: 3.725788E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 853.930 | TFLOPs: 29.89 | 7: iteration 5940/ 21553 | consumed samples: 1520640 | consumed tokens: 3114270720 | elapsed time per iteration (s): 0.30 | learning rate: 1.699E-04 | global batch size: 256 | lm loss: 3.722137E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.129 | TFLOPs: 30.32 | 7: iteration 5950/ 21553 | consumed samples: 1523200 | consumed tokens: 3119513600 | elapsed time per iteration (s): 0.30 | learning rate: 1.698E-04 | global batch size: 256 | lm loss: 3.727018E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.086 | TFLOPs: 30.32 | 7: iteration 5960/ 21553 | consumed samples: 1525760 | consumed tokens: 3124756480 | elapsed time per iteration (s): 0.30 | learning rate: 1.697E-04 | global batch size: 256 | lm loss: 3.730131E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.943 | TFLOPs: 30.31 | 7: iteration 5970/ 21553 | consumed samples: 1528320 | consumed tokens: 3129999360 | elapsed time per iteration (s): 0.30 | learning rate: 1.696E-04 | global batch size: 256 | lm loss: 3.722199E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.974 | TFLOPs: 30.32 | 7: iteration 5980/ 21553 | consumed samples: 1530880 | consumed tokens: 3135242240 | elapsed time per iteration (s): 0.30 | learning rate: 1.695E-04 | global batch size: 256 | lm loss: 3.725134E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 855.235 | TFLOPs: 29.94 | 7: iteration 5990/ 21553 | consumed samples: 1533440 | consumed tokens: 3140485120 | elapsed time per iteration (s): 0.30 | learning rate: 1.694E-04 | global batch size: 256 | lm loss: 3.717149E+00 | grad norm: 0.482 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.994 | TFLOPs: 30.32 | 0: [2023-03-17 00:25:40,679] [INFO] [logging.py:68:log_dist] [Rank 0] step=6000, skipped=0, lr=[0.00016928570742907802, 0.00016928570742907802, 0.00016928570742907802], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 6000/ 21553 | consumed samples: 1536000 | consumed tokens: 3145728000 | elapsed time per iteration (s): 0.30 | learning rate: 1.693E-04 | global batch size: 256 | lm loss: 3.721169E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.326 | TFLOPs: 30.33 | 0: steps: 6000 loss: 3.7337 iter time (s): 0.294 samples/sec: 871.381 7: ----------------------------------------------------------------------------------------------- 7: validation loss at iteration 6000 | lm loss value: 3.900856E+00 | lm loss PPL: 4.944473E+01 | 7: ----------------------------------------------------------------------------------------------- 0: saving checkpoint at iteration 6000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:25:40,801] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step6000 is begin to save! 0: [2023-03-17 00:25:40,804] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:25:40,889] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:25:40,889] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:25:40,905] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:25:40,905] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:25:40,920] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:25:40,920] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:25:40,934] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:25:40,935] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:25:40,949] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:25:40,950] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:25:40,964] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:25:40,965] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:25:40,979] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:25:40,980] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:25:40,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:25:40,994] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:25:41,009] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:25:41,009] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:25:41,024] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:25:41,024] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:25:41,039] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:25:41,039] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:25:41,054] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:25:41,054] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:25:41,069] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:25:41,069] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:25:41,084] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:25:41,084] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:25:41,099] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:25:41,099] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:25:41,114] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:25:41,114] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:25:41,115] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:25:41,116] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step6000/mp_rank_00_model_states.pt 0: [2023-03-17 00:25:41,116] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:25:41,118] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:25:41,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:25:41,168] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:25:41,170] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:25:41,170] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:25:41,170] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:25:41,170] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 00:25:41,170] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 3: [2023-03-17 00:25:41,170] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 0: [2023-03-17 00:25:41,170] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 3: [2023-03-17 00:25:41,170] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 0: [2023-03-17 00:25:41,170] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 4: [2023-03-17 00:25:41,171] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:25:41,171] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 1: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:25:41,172] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 4: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:25:41,172] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 00:25:41,172] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 5: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 5: [2023-03-17 00:25:41,172] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 7: [2023-03-17 00:25:41,169] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,169] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 00:25:41,169] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 3: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,170] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,170] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 00:25:41,170] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 7: [2023-03-17 00:25:41,171] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,171] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 3: [2023-03-17 00:25:41,172] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,171] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 3: [2023-03-17 00:25:41,172] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 6: [2023-03-17 00:25:41,172] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,172] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,173] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 6: [2023-03-17 00:25:41,173] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 5: [2023-03-17 00:25:41,173] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:25:41,173] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:25:41,173] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 2: [2023-03-17 00:25:41,173] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 5: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 2: [2023-03-17 00:25:41,173] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 2: [2023-03-17 00:25:41,173] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:25:41,174] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 4: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:25:41,174] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 2: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:25:41,174] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 00:25:41,174] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 2: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 6: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:25:41,174] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 4: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:25:41,175] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 5: [2023-03-17 00:25:41,175] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 0: [2023-03-17 00:25:41,175] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 4: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 5: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 0: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 0: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:25:41,175] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 2: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:25:41,175] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 3: [2023-03-17 00:25:41,175] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 2: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 3: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 1: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:25:41,176] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 1: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:25:41,176] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 1: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:25:41,176] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:25:41,176] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 3: [2023-03-17 00:25:41,177] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 1: [2023-03-17 00:25:41,177] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 1: [2023-03-17 00:25:41,177] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 3: [2023-03-17 00:25:41,177] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 7: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,176] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 7: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,176] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,176] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 00:25:41,177] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 7: [2023-03-17 00:25:41,177] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:25:41,177] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 4: [2023-03-17 00:25:41,177] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:25:41,177] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 00:25:41,177] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 3: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:25:41,178] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 2: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:25:41,174] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 6: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:25:41,174] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,174] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 6: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:25:41,175] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,175] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 4: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:25:41,178] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,178] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,178] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 6: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 4: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 5: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:25:41,178] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 3: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 3: [2023-03-17 00:25:41,178] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 5: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:25:41,179] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 00:25:41,179] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 2: [2023-03-17 00:25:41,178] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 2: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:25:41,178] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 2: [2023-03-17 00:25:41,178] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:25:41,178] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 1: [2023-03-17 00:25:41,179] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:25:41,179] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:25:41,179] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:25:41,179] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:25:41,179] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,179] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 00:25:41,179] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 1: [2023-03-17 00:25:41,179] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 1: [2023-03-17 00:25:41,179] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 1: [2023-03-17 00:25:41,179] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:25:41,179] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:25:41,179] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 5: [2023-03-17 00:25:41,181] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:25:41,181] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:25:41,181] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 5: [2023-03-17 00:25:41,181] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:25:41,181] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:25:41,181] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 5: [2023-03-17 00:25:41,181] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:25:41,181] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:25:41,181] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 3: [2023-03-17 00:25:41,181] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 5: [2023-03-17 00:25:41,181] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 3: [2023-03-17 00:25:41,181] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 0: [2023-03-17 00:25:41,183] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:25:41,183] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 00:25:41,184] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 4: [2023-03-17 00:25:41,184] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:25:41,184] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:25:41,184] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 0: [2023-03-17 00:25:41,188] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:25:41,188] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:25:41,188] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:25:41,188] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 00:25:41,188] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 0: [2023-03-17 00:25:41,188] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 0: [2023-03-17 00:25:41,194] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:25:41,194] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 7: [2023-03-17 00:25:41,183] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,184] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 00:25:41,184] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 7: [2023-03-17 00:25:41,197] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:25:41,197] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step6000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:25:41,197] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step6000 is ready now! 0: successfully saved checkpoint at iteration 6000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 402.21 7: iteration 6010/ 21553 | consumed samples: 1538560 | consumed tokens: 3150970880 | elapsed time per iteration (s): 0.35 | learning rate: 1.692E-04 | global batch size: 256 | lm loss: 3.712474E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 737.849 | TFLOPs: 25.83 | 7: iteration 6020/ 21553 | consumed samples: 1541120 | consumed tokens: 3156213760 | elapsed time per iteration (s): 0.30 | learning rate: 1.691E-04 | global batch size: 256 | lm loss: 3.709961E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.837 | TFLOPs: 30.35 | 7: iteration 6030/ 21553 | consumed samples: 1543680 | consumed tokens: 3161456640 | elapsed time per iteration (s): 0.30 | learning rate: 1.690E-04 | global batch size: 256 | lm loss: 3.708266E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.070 | TFLOPs: 30.35 | 7: iteration 6040/ 21553 | consumed samples: 1546240 | consumed tokens: 3166699520 | elapsed time per iteration (s): 0.30 | learning rate: 1.689E-04 | global batch size: 256 | lm loss: 3.712395E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.489 | TFLOPs: 30.33 | 7: iteration 6050/ 21553 | consumed samples: 1548800 | consumed tokens: 3171942400 | elapsed time per iteration (s): 0.30 | learning rate: 1.688E-04 | global batch size: 256 | lm loss: 3.715368E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.168 | TFLOPs: 30.32 | 7: iteration 6060/ 21553 | consumed samples: 1551360 | consumed tokens: 3177185280 | elapsed time per iteration (s): 0.30 | learning rate: 1.687E-04 | global batch size: 256 | lm loss: 3.718854E+00 | grad norm: 0.371 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.443 | TFLOPs: 30.33 | 7: iteration 6070/ 21553 | consumed samples: 1553920 | consumed tokens: 3182428160 | elapsed time per iteration (s): 0.30 | learning rate: 1.686E-04 | global batch size: 256 | lm loss: 3.709176E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.403 | TFLOPs: 30.33 | 7: iteration 6080/ 21553 | consumed samples: 1556480 | consumed tokens: 3187671040 | elapsed time per iteration (s): 0.30 | learning rate: 1.685E-04 | global batch size: 256 | lm loss: 3.716927E+00 | grad norm: 0.388 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 860.784 | TFLOPs: 30.13 | 7: iteration 6090/ 21553 | consumed samples: 1559040 | consumed tokens: 3192913920 | elapsed time per iteration (s): 0.30 | learning rate: 1.684E-04 | global batch size: 256 | lm loss: 3.710141E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.102 | TFLOPs: 30.04 | 7: iteration 6100/ 21553 | consumed samples: 1561600 | consumed tokens: 3198156800 | elapsed time per iteration (s): 0.30 | learning rate: 1.683E-04 | global batch size: 256 | lm loss: 3.714998E+00 | grad norm: 0.372 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.518 | TFLOPs: 30.30 | 7: iteration 6110/ 21553 | consumed samples: 1564160 | consumed tokens: 3203399680 | elapsed time per iteration (s): 0.30 | learning rate: 1.682E-04 | global batch size: 256 | lm loss: 3.712304E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.379 | TFLOPs: 30.29 | 7: iteration 6120/ 21553 | consumed samples: 1566720 | consumed tokens: 3208642560 | elapsed time per iteration (s): 0.30 | learning rate: 1.681E-04 | global batch size: 256 | lm loss: 3.708679E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.764 | TFLOPs: 30.20 | 7: iteration 6130/ 21553 | consumed samples: 1569280 | consumed tokens: 3213885440 | elapsed time per iteration (s): 0.30 | learning rate: 1.680E-04 | global batch size: 256 | lm loss: 3.711931E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.252 | TFLOPs: 30.33 | 7: iteration 6140/ 21553 | consumed samples: 1571840 | consumed tokens: 3219128320 | elapsed time per iteration (s): 0.30 | learning rate: 1.679E-04 | global batch size: 256 | lm loss: 3.705112E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.982 | TFLOPs: 30.32 | 7: iteration 6150/ 21553 | consumed samples: 1574400 | consumed tokens: 3224371200 | elapsed time per iteration (s): 0.30 | learning rate: 1.678E-04 | global batch size: 256 | lm loss: 3.704380E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.734 | TFLOPs: 30.31 | 7: iteration 6160/ 21553 | consumed samples: 1576960 | consumed tokens: 3229614080 | elapsed time per iteration (s): 0.30 | learning rate: 1.677E-04 | global batch size: 256 | lm loss: 3.707771E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.638 | TFLOPs: 30.06 | 7: iteration 6170/ 21553 | consumed samples: 1579520 | consumed tokens: 3234856960 | elapsed time per iteration (s): 0.30 | learning rate: 1.676E-04 | global batch size: 256 | lm loss: 3.707170E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.650 | TFLOPs: 30.34 | 7: iteration 6180/ 21553 | consumed samples: 1582080 | consumed tokens: 3240099840 | elapsed time per iteration (s): 0.30 | learning rate: 1.675E-04 | global batch size: 256 | lm loss: 3.708372E+00 | grad norm: 0.370 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.867 | TFLOPs: 30.35 | 7: iteration 6190/ 21553 | consumed samples: 1584640 | consumed tokens: 3245342720 | elapsed time per iteration (s): 0.30 | learning rate: 1.674E-04 | global batch size: 256 | lm loss: 3.705286E+00 | grad norm: 0.377 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.807 | TFLOPs: 30.31 | 7: iteration 6200/ 21553 | consumed samples: 1587200 | consumed tokens: 3250585600 | elapsed time per iteration (s): 0.30 | learning rate: 1.673E-04 | global batch size: 256 | lm loss: 3.706959E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.559 | TFLOPs: 30.30 | 7: iteration 6210/ 21553 | consumed samples: 1589760 | consumed tokens: 3255828480 | elapsed time per iteration (s): 0.30 | learning rate: 1.672E-04 | global batch size: 256 | lm loss: 3.706635E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.080 | TFLOPs: 30.32 | 7: iteration 6220/ 21553 | consumed samples: 1592320 | consumed tokens: 3261071360 | elapsed time per iteration (s): 0.30 | learning rate: 1.671E-04 | global batch size: 256 | lm loss: 3.705392E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.588 | TFLOPs: 30.30 | 7: iteration 6230/ 21553 | consumed samples: 1594880 | consumed tokens: 3266314240 | elapsed time per iteration (s): 0.30 | learning rate: 1.670E-04 | global batch size: 256 | lm loss: 3.711074E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.652 | TFLOPs: 30.30 | 7: iteration 6240/ 21553 | consumed samples: 1597440 | consumed tokens: 3271557120 | elapsed time per iteration (s): 0.30 | learning rate: 1.669E-04 | global batch size: 256 | lm loss: 3.708056E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.819 | TFLOPs: 30.31 | 7: iteration 6250/ 21553 | consumed samples: 1600000 | consumed tokens: 3276800000 | elapsed time per iteration (s): 0.30 | learning rate: 1.668E-04 | global batch size: 256 | lm loss: 3.701576E+00 | grad norm: 0.381 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.181 | TFLOPs: 30.32 | 7: iteration 6260/ 21553 | consumed samples: 1602560 | consumed tokens: 3282042880 | elapsed time per iteration (s): 0.30 | learning rate: 1.667E-04 | global batch size: 256 | lm loss: 3.693425E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.913 | TFLOPs: 30.31 | 7: iteration 6270/ 21553 | consumed samples: 1605120 | consumed tokens: 3287285760 | elapsed time per iteration (s): 0.30 | learning rate: 1.665E-04 | global batch size: 256 | lm loss: 3.698648E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.917 | TFLOPs: 30.31 | 7: iteration 6280/ 21553 | consumed samples: 1607680 | consumed tokens: 3292528640 | elapsed time per iteration (s): 0.30 | learning rate: 1.664E-04 | global batch size: 256 | lm loss: 3.703384E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.673 | TFLOPs: 30.30 | 7: iteration 6290/ 21553 | consumed samples: 1610240 | consumed tokens: 3297771520 | elapsed time per iteration (s): 0.30 | learning rate: 1.663E-04 | global batch size: 256 | lm loss: 3.699393E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.173 | TFLOPs: 30.32 | 7: iteration 6300/ 21553 | consumed samples: 1612800 | consumed tokens: 3303014400 | elapsed time per iteration (s): 0.30 | learning rate: 1.662E-04 | global batch size: 256 | lm loss: 3.703036E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.049 | TFLOPs: 30.32 | 7: iteration 6310/ 21553 | consumed samples: 1615360 | consumed tokens: 3308257280 | elapsed time per iteration (s): 0.30 | learning rate: 1.661E-04 | global batch size: 256 | lm loss: 3.706093E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.220 | TFLOPs: 30.32 | 7: iteration 6320/ 21553 | consumed samples: 1617920 | consumed tokens: 3313500160 | elapsed time per iteration (s): 0.30 | learning rate: 1.660E-04 | global batch size: 256 | lm loss: 3.696106E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.671 | TFLOPs: 30.34 | 7: iteration 6330/ 21553 | consumed samples: 1620480 | consumed tokens: 3318743040 | elapsed time per iteration (s): 0.30 | learning rate: 1.659E-04 | global batch size: 256 | lm loss: 3.696181E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.442 | TFLOPs: 30.33 | 7: iteration 6340/ 21553 | consumed samples: 1623040 | consumed tokens: 3323985920 | elapsed time per iteration (s): 0.30 | learning rate: 1.658E-04 | global batch size: 256 | lm loss: 3.690911E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.345 | TFLOPs: 30.33 | 7: iteration 6350/ 21553 | consumed samples: 1625600 | consumed tokens: 3329228800 | elapsed time per iteration (s): 0.30 | learning rate: 1.657E-04 | global batch size: 256 | lm loss: 3.694274E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.791 | TFLOPs: 30.34 | 7: iteration 6360/ 21553 | consumed samples: 1628160 | consumed tokens: 3334471680 | elapsed time per iteration (s): 0.30 | learning rate: 1.656E-04 | global batch size: 256 | lm loss: 3.688482E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.191 | TFLOPs: 30.32 | 7: iteration 6370/ 21553 | consumed samples: 1630720 | consumed tokens: 3339714560 | elapsed time per iteration (s): 0.30 | learning rate: 1.655E-04 | global batch size: 256 | lm loss: 3.697244E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.098 | TFLOPs: 30.32 | 7: iteration 6380/ 21553 | consumed samples: 1633280 | consumed tokens: 3344957440 | elapsed time per iteration (s): 0.30 | learning rate: 1.654E-04 | global batch size: 256 | lm loss: 3.692172E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.122 | TFLOPs: 30.32 | 7: iteration 6390/ 21553 | consumed samples: 1635840 | consumed tokens: 3350200320 | elapsed time per iteration (s): 0.30 | learning rate: 1.653E-04 | global batch size: 256 | lm loss: 3.683580E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.386 | TFLOPs: 30.33 | 7: iteration 6400/ 21553 | consumed samples: 1638400 | consumed tokens: 3355443200 | elapsed time per iteration (s): 0.30 | learning rate: 1.652E-04 | global batch size: 256 | lm loss: 3.689849E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.725 | TFLOPs: 30.34 | 7: iteration 6410/ 21553 | consumed samples: 1640960 | consumed tokens: 3360686080 | elapsed time per iteration (s): 0.30 | learning rate: 1.651E-04 | global batch size: 256 | lm loss: 3.693872E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.310 | TFLOPs: 30.33 | 7: iteration 6420/ 21553 | consumed samples: 1643520 | consumed tokens: 3365928960 | elapsed time per iteration (s): 0.30 | learning rate: 1.650E-04 | global batch size: 256 | lm loss: 3.685575E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.830 | TFLOPs: 30.31 | 7: iteration 6430/ 21553 | consumed samples: 1646080 | consumed tokens: 3371171840 | elapsed time per iteration (s): 0.30 | learning rate: 1.649E-04 | global batch size: 256 | lm loss: 3.692246E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 840.510 | TFLOPs: 29.42 | 7: iteration 6440/ 21553 | consumed samples: 1648640 | consumed tokens: 3376414720 | elapsed time per iteration (s): 0.30 | learning rate: 1.648E-04 | global batch size: 256 | lm loss: 3.690814E+00 | grad norm: 0.386 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.558 | TFLOPs: 30.34 | 7: iteration 6450/ 21553 | consumed samples: 1651200 | consumed tokens: 3381657600 | elapsed time per iteration (s): 0.30 | learning rate: 1.647E-04 | global batch size: 256 | lm loss: 3.694168E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.547 | TFLOPs: 30.34 | 7: iteration 6460/ 21553 | consumed samples: 1653760 | consumed tokens: 3386900480 | elapsed time per iteration (s): 0.30 | learning rate: 1.646E-04 | global batch size: 256 | lm loss: 3.688998E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.274 | TFLOPs: 30.33 | 7: iteration 6470/ 21553 | consumed samples: 1656320 | consumed tokens: 3392143360 | elapsed time per iteration (s): 0.30 | learning rate: 1.645E-04 | global batch size: 256 | lm loss: 3.683297E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.683 | TFLOPs: 30.31 | 7: iteration 6480/ 21553 | consumed samples: 1658880 | consumed tokens: 3397386240 | elapsed time per iteration (s): 0.30 | learning rate: 1.644E-04 | global batch size: 256 | lm loss: 3.686221E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.543 | TFLOPs: 30.30 | 7: iteration 6490/ 21553 | consumed samples: 1661440 | consumed tokens: 3402629120 | elapsed time per iteration (s): 0.30 | learning rate: 1.643E-04 | global batch size: 256 | lm loss: 3.700274E+00 | grad norm: 0.458 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.433 | TFLOPs: 30.30 | 7: iteration 6500/ 21553 | consumed samples: 1664000 | consumed tokens: 3407872000 | elapsed time per iteration (s): 0.30 | learning rate: 1.641E-04 | global batch size: 256 | lm loss: 3.682690E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.553 | TFLOPs: 30.30 | 7: iteration 6510/ 21553 | consumed samples: 1666560 | consumed tokens: 3413114880 | elapsed time per iteration (s): 0.30 | learning rate: 1.640E-04 | global batch size: 256 | lm loss: 3.679615E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.407 | TFLOPs: 30.30 | 7: iteration 6520/ 21553 | consumed samples: 1669120 | consumed tokens: 3418357760 | elapsed time per iteration (s): 0.30 | learning rate: 1.639E-04 | global batch size: 256 | lm loss: 3.692976E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.865 | TFLOPs: 30.31 | 7: iteration 6530/ 21553 | consumed samples: 1671680 | consumed tokens: 3423600640 | elapsed time per iteration (s): 0.30 | learning rate: 1.638E-04 | global batch size: 256 | lm loss: 3.681854E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.664 | TFLOPs: 30.30 | 7: iteration 6540/ 21553 | consumed samples: 1674240 | consumed tokens: 3428843520 | elapsed time per iteration (s): 0.30 | learning rate: 1.637E-04 | global batch size: 256 | lm loss: 3.682634E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.510 | TFLOPs: 30.30 | 7: iteration 6550/ 21553 | consumed samples: 1676800 | consumed tokens: 3434086400 | elapsed time per iteration (s): 0.30 | learning rate: 1.636E-04 | global batch size: 256 | lm loss: 3.681161E+00 | grad norm: 0.384 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.356 | TFLOPs: 30.29 | 7: iteration 6560/ 21553 | consumed samples: 1679360 | consumed tokens: 3439329280 | elapsed time per iteration (s): 0.30 | learning rate: 1.635E-04 | global batch size: 256 | lm loss: 3.684070E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.207 | TFLOPs: 30.29 | 7: iteration 6570/ 21553 | consumed samples: 1681920 | consumed tokens: 3444572160 | elapsed time per iteration (s): 0.30 | learning rate: 1.634E-04 | global batch size: 256 | lm loss: 3.686751E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.456 | TFLOPs: 30.30 | 7: iteration 6580/ 21553 | consumed samples: 1684480 | consumed tokens: 3449815040 | elapsed time per iteration (s): 0.30 | learning rate: 1.633E-04 | global batch size: 256 | lm loss: 3.677476E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.660 | TFLOPs: 30.30 | 7: iteration 6590/ 21553 | consumed samples: 1687040 | consumed tokens: 3455057920 | elapsed time per iteration (s): 0.30 | learning rate: 1.632E-04 | global batch size: 256 | lm loss: 3.686499E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.460 | TFLOPs: 30.30 | 7: iteration 6600/ 21553 | consumed samples: 1689600 | consumed tokens: 3460300800 | elapsed time per iteration (s): 0.30 | learning rate: 1.631E-04 | global batch size: 256 | lm loss: 3.681258E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.521 | TFLOPs: 30.30 | 7: iteration 6610/ 21553 | consumed samples: 1692160 | consumed tokens: 3465543680 | elapsed time per iteration (s): 0.30 | learning rate: 1.630E-04 | global batch size: 256 | lm loss: 3.680004E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.571 | TFLOPs: 30.30 | 7: iteration 6620/ 21553 | consumed samples: 1694720 | consumed tokens: 3470786560 | elapsed time per iteration (s): 0.30 | learning rate: 1.629E-04 | global batch size: 256 | lm loss: 3.676452E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.579 | TFLOPs: 30.30 | 7: iteration 6630/ 21553 | consumed samples: 1697280 | consumed tokens: 3476029440 | elapsed time per iteration (s): 0.30 | learning rate: 1.628E-04 | global batch size: 256 | lm loss: 3.680541E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.354 | TFLOPs: 30.29 | 7: iteration 6640/ 21553 | consumed samples: 1699840 | consumed tokens: 3481272320 | elapsed time per iteration (s): 0.30 | learning rate: 1.627E-04 | global batch size: 256 | lm loss: 3.680004E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.686 | TFLOPs: 30.31 | 7: iteration 6650/ 21553 | consumed samples: 1702400 | consumed tokens: 3486515200 | elapsed time per iteration (s): 0.30 | learning rate: 1.625E-04 | global batch size: 256 | lm loss: 3.669538E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.562 | TFLOPs: 30.30 | 7: iteration 6660/ 21553 | consumed samples: 1704960 | consumed tokens: 3491758080 | elapsed time per iteration (s): 0.30 | learning rate: 1.624E-04 | global batch size: 256 | lm loss: 3.674881E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.407 | TFLOPs: 30.33 | 7: iteration 6670/ 21553 | consumed samples: 1707520 | consumed tokens: 3497000960 | elapsed time per iteration (s): 0.30 | learning rate: 1.623E-04 | global batch size: 256 | lm loss: 3.679854E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.519 | TFLOPs: 30.33 | 7: iteration 6680/ 21553 | consumed samples: 1710080 | consumed tokens: 3502243840 | elapsed time per iteration (s): 0.30 | learning rate: 1.622E-04 | global batch size: 256 | lm loss: 3.671127E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.312 | TFLOPs: 30.33 | 7: iteration 6690/ 21553 | consumed samples: 1712640 | consumed tokens: 3507486720 | elapsed time per iteration (s): 0.30 | learning rate: 1.621E-04 | global batch size: 256 | lm loss: 3.673207E+00 | grad norm: 0.386 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.222 | TFLOPs: 30.32 | 7: iteration 6700/ 21553 | consumed samples: 1715200 | consumed tokens: 3512729600 | elapsed time per iteration (s): 0.30 | learning rate: 1.620E-04 | global batch size: 256 | lm loss: 3.677798E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.107 | TFLOPs: 30.32 | 7: iteration 6710/ 21553 | consumed samples: 1717760 | consumed tokens: 3517972480 | elapsed time per iteration (s): 0.30 | learning rate: 1.619E-04 | global batch size: 256 | lm loss: 3.669930E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.819 | TFLOPs: 30.31 | 7: iteration 6720/ 21553 | consumed samples: 1720320 | consumed tokens: 3523215360 | elapsed time per iteration (s): 0.30 | learning rate: 1.618E-04 | global batch size: 256 | lm loss: 3.679343E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.501 | TFLOPs: 30.33 | 7: iteration 6730/ 21553 | consumed samples: 1722880 | consumed tokens: 3528458240 | elapsed time per iteration (s): 0.30 | learning rate: 1.617E-04 | global batch size: 256 | lm loss: 3.670502E+00 | grad norm: 0.387 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.886 | TFLOPs: 30.35 | 7: iteration 6740/ 21553 | consumed samples: 1725440 | consumed tokens: 3533701120 | elapsed time per iteration (s): 0.30 | learning rate: 1.616E-04 | global batch size: 256 | lm loss: 3.669292E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.455 | TFLOPs: 30.33 | 7: iteration 6750/ 21553 | consumed samples: 1728000 | consumed tokens: 3538944000 | elapsed time per iteration (s): 0.30 | learning rate: 1.615E-04 | global batch size: 256 | lm loss: 3.679406E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.700 | TFLOPs: 30.34 | 7: iteration 6760/ 21553 | consumed samples: 1730560 | consumed tokens: 3544186880 | elapsed time per iteration (s): 0.30 | learning rate: 1.614E-04 | global batch size: 256 | lm loss: 3.661719E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.287 | TFLOPs: 30.05 | 7: iteration 6770/ 21553 | consumed samples: 1733120 | consumed tokens: 3549429760 | elapsed time per iteration (s): 0.30 | learning rate: 1.612E-04 | global batch size: 256 | lm loss: 3.673714E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.594 | TFLOPs: 30.30 | 7: iteration 6780/ 21553 | consumed samples: 1735680 | consumed tokens: 3554672640 | elapsed time per iteration (s): 0.30 | learning rate: 1.611E-04 | global batch size: 256 | lm loss: 3.676040E+00 | grad norm: 0.388 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.027 | TFLOPs: 30.32 | 7: iteration 6790/ 21553 | consumed samples: 1738240 | consumed tokens: 3559915520 | elapsed time per iteration (s): 0.30 | learning rate: 1.610E-04 | global batch size: 256 | lm loss: 3.675249E+00 | grad norm: 0.388 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.107 | TFLOPs: 30.32 | 7: iteration 6800/ 21553 | consumed samples: 1740800 | consumed tokens: 3565158400 | elapsed time per iteration (s): 0.30 | learning rate: 1.609E-04 | global batch size: 256 | lm loss: 3.661789E+00 | grad norm: 0.384 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.263 | TFLOPs: 30.33 | 7: iteration 6810/ 21553 | consumed samples: 1743360 | consumed tokens: 3570401280 | elapsed time per iteration (s): 0.30 | learning rate: 1.608E-04 | global batch size: 256 | lm loss: 3.674067E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.338 | TFLOPs: 30.33 | 7: iteration 6820/ 21553 | consumed samples: 1745920 | consumed tokens: 3575644160 | elapsed time per iteration (s): 0.30 | learning rate: 1.607E-04 | global batch size: 256 | lm loss: 3.668835E+00 | grad norm: 0.383 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.551 | TFLOPs: 30.34 | 7: iteration 6830/ 21553 | consumed samples: 1748480 | consumed tokens: 3580887040 | elapsed time per iteration (s): 0.30 | learning rate: 1.606E-04 | global batch size: 256 | lm loss: 3.663457E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.377 | TFLOPs: 30.33 | 7: iteration 6840/ 21553 | consumed samples: 1751040 | consumed tokens: 3586129920 | elapsed time per iteration (s): 0.30 | learning rate: 1.605E-04 | global batch size: 256 | lm loss: 3.661073E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.363 | TFLOPs: 30.33 | 7: iteration 6850/ 21553 | consumed samples: 1753600 | consumed tokens: 3591372800 | elapsed time per iteration (s): 0.30 | learning rate: 1.604E-04 | global batch size: 256 | lm loss: 3.670274E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.548 | TFLOPs: 30.34 | 7: iteration 6860/ 21553 | consumed samples: 1756160 | consumed tokens: 3596615680 | elapsed time per iteration (s): 0.30 | learning rate: 1.603E-04 | global batch size: 256 | lm loss: 3.662763E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.418 | TFLOPs: 30.33 | 7: iteration 6870/ 21553 | consumed samples: 1758720 | consumed tokens: 3601858560 | elapsed time per iteration (s): 0.30 | learning rate: 1.601E-04 | global batch size: 256 | lm loss: 3.662974E+00 | grad norm: 0.387 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.565 | TFLOPs: 30.34 | 7: iteration 6880/ 21553 | consumed samples: 1761280 | consumed tokens: 3607101440 | elapsed time per iteration (s): 0.30 | learning rate: 1.600E-04 | global batch size: 256 | lm loss: 3.656705E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.397 | TFLOPs: 30.33 | 7: iteration 6890/ 21553 | consumed samples: 1763840 | consumed tokens: 3612344320 | elapsed time per iteration (s): 0.30 | learning rate: 1.599E-04 | global batch size: 256 | lm loss: 3.661686E+00 | grad norm: 0.385 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.790 | TFLOPs: 30.34 | 7: iteration 6900/ 21553 | consumed samples: 1766400 | consumed tokens: 3617587200 | elapsed time per iteration (s): 0.30 | learning rate: 1.598E-04 | global batch size: 256 | lm loss: 3.659167E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.776 | TFLOPs: 30.34 | 7: iteration 6910/ 21553 | consumed samples: 1768960 | consumed tokens: 3622830080 | elapsed time per iteration (s): 0.30 | learning rate: 1.597E-04 | global batch size: 256 | lm loss: 3.658815E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.565 | TFLOPs: 30.34 | 7: iteration 6920/ 21553 | consumed samples: 1771520 | consumed tokens: 3628072960 | elapsed time per iteration (s): 0.30 | learning rate: 1.596E-04 | global batch size: 256 | lm loss: 3.653032E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.098 | TFLOPs: 30.32 | 7: iteration 6930/ 21553 | consumed samples: 1774080 | consumed tokens: 3633315840 | elapsed time per iteration (s): 0.30 | learning rate: 1.595E-04 | global batch size: 256 | lm loss: 3.658111E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.496 | TFLOPs: 30.33 | 7: iteration 6940/ 21553 | consumed samples: 1776640 | consumed tokens: 3638558720 | elapsed time per iteration (s): 0.30 | learning rate: 1.594E-04 | global batch size: 256 | lm loss: 3.660674E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.765 | TFLOPs: 30.34 | 7: iteration 6950/ 21553 | consumed samples: 1779200 | consumed tokens: 3643801600 | elapsed time per iteration (s): 0.30 | learning rate: 1.593E-04 | global batch size: 256 | lm loss: 3.662992E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.650 | TFLOPs: 30.34 | 7: iteration 6960/ 21553 | consumed samples: 1781760 | consumed tokens: 3649044480 | elapsed time per iteration (s): 0.30 | learning rate: 1.592E-04 | global batch size: 256 | lm loss: 3.662474E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.151 | TFLOPs: 30.32 | 7: iteration 6970/ 21553 | consumed samples: 1784320 | consumed tokens: 3654287360 | elapsed time per iteration (s): 0.30 | learning rate: 1.590E-04 | global batch size: 256 | lm loss: 3.660831E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.152 | TFLOPs: 30.32 | 7: iteration 6980/ 21553 | consumed samples: 1786880 | consumed tokens: 3659530240 | elapsed time per iteration (s): 0.30 | learning rate: 1.589E-04 | global batch size: 256 | lm loss: 3.653612E+00 | grad norm: 0.388 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.572 | TFLOPs: 30.34 | 7: iteration 6990/ 21553 | consumed samples: 1789440 | consumed tokens: 3664773120 | elapsed time per iteration (s): 0.30 | learning rate: 1.588E-04 | global batch size: 256 | lm loss: 3.658204E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.409 | TFLOPs: 30.33 | 7: iteration 7000/ 21553 | consumed samples: 1792000 | consumed tokens: 3670016000 | elapsed time per iteration (s): 0.30 | learning rate: 1.587E-04 | global batch size: 256 | lm loss: 3.660359E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.994 | TFLOPs: 30.32 | 7: ----------------------------------------------------------------------------------------------- 7: validation loss at iteration 7000 | lm loss value: 3.900835E+00 | lm loss PPL: 4.944373E+01 | 7: ----------------------------------------------------------------------------------------------- 0: saving checkpoint at iteration 7000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:30:37,067] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step7000 is begin to save! 0: [2023-03-17 00:30:37,070] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:30:37,155] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:30:37,155] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:30:37,171] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:30:37,171] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:30:37,188] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:30:37,188] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:30:37,203] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:30:37,203] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:30:37,218] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:30:37,218] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:30:37,233] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:30:37,233] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:30:37,247] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:30:37,248] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:30:37,262] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:30:37,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:30:37,277] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:30:37,277] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:30:37,292] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:30:37,292] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:30:37,306] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:30:37,307] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:30:37,321] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:30:37,322] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:30:37,336] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:30:37,336] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:30:37,351] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:30:37,351] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:30:37,366] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:30:37,366] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:30:37,381] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:30:37,381] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:30:37,382] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:30:37,383] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step7000/mp_rank_00_model_states.pt 0: [2023-03-17 00:30:37,383] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:30:37,385] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:30:37,401] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:30:37,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:30:37,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:30:37,434] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,435] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,435] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:30:37,435] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 3: [2023-03-17 00:30:37,435] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,435] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:30:37,435] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 5: [2023-03-17 00:30:37,436] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:30:37,436] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 00:30:37,436] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 5: [2023-03-17 00:30:37,436] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:30:37,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 1: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:30:37,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 0: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:30:37,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 0: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:30:37,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 4: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:30:37,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 4: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:30:37,437] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 00:30:37,437] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 4: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:30:37,438] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 2: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:30:37,438] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 0: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:30:37,438] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 5: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:30:37,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:30:37,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:30:37,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: [2023-03-17 00:30:37,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 3: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 5: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:30:37,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:30:37,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 5: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 5: [2023-03-17 00:30:37,440] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:30:37,440] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:30:37,440] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 6: [2023-03-17 00:30:37,433] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 3: [2023-03-17 00:30:37,440] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:30:37,434] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 6: [2023-03-17 00:30:37,435] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,440] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 6: [2023-03-17 00:30:37,435] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:30:37,436] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 6: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,440] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 6: [2023-03-17 00:30:37,438] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 6: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:30:37,439] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 3: [2023-03-17 00:30:37,440] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,440] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 00:30:37,441] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 0: [2023-03-17 00:30:37,441] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:30:37,442] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 0: [2023-03-17 00:30:37,442] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 3: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,442] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 0: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 3: [2023-03-17 00:30:37,442] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 4: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 4: [2023-03-17 00:30:37,442] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 1: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:30:37,443] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 1: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:30:37,443] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 1: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:30:37,443] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 2: [2023-03-17 00:30:37,438] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:30:37,438] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 2: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:30:37,440] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:30:37,440] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 2: [2023-03-17 00:30:37,441] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:30:37,441] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 00:30:37,441] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 2: [2023-03-17 00:30:37,441] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:30:37,441] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:30:37,441] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 2: [2023-03-17 00:30:37,441] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:30:37,441] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:30:37,441] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 2: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:30:37,443] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 0: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:30:37,444] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 3: [2023-03-17 00:30:37,444] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 0: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 0: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 0: [2023-03-17 00:30:37,444] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 5: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:30:37,445] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 5: [2023-03-17 00:30:37,445] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 6: [2023-03-17 00:30:37,442] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:30:37,443] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 6: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:30:37,443] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:30:37,443] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 6: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 6: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:30:37,444] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 2: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:30:37,445] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:30:37,445] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 2: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 1: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:30:37,445] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:30:37,445] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 00:30:37,446] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 1: [2023-03-17 00:30:37,446] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 1: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:30:37,446] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:30:37,446] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:30:37,446] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:30:37,446] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 1: [2023-03-17 00:30:37,446] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: [2023-03-17 00:30:37,439] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: [2023-03-17 00:30:37,440] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:30:37,440] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 00:30:37,440] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:30:37,443] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:30:37,443] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:30:37,444] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:30:37,444] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: [2023-03-17 00:30:37,444] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:30:37,444] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 00:30:37,445] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 4: [2023-03-17 00:30:37,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:30:37,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 00:30:37,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 4: [2023-03-17 00:30:37,450] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:30:37,450] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:30:37,450] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 4: [2023-03-17 00:30:37,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 0: successfully saved checkpoint at iteration 7000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 389.74 4: [2023-03-17 00:30:37,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:30:37,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 4: [2023-03-17 00:30:37,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:30:37,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step7000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:30:37,453] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step7000 is ready now! 7: iteration 7010/ 21553 | consumed samples: 1794560 | consumed tokens: 3675258880 | elapsed time per iteration (s): 0.35 | learning rate: 1.586E-04 | global batch size: 256 | lm loss: 3.655769E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 741.117 | TFLOPs: 25.94 | 7: iteration 7020/ 21553 | consumed samples: 1797120 | consumed tokens: 3680501760 | elapsed time per iteration (s): 0.30 | learning rate: 1.585E-04 | global batch size: 256 | lm loss: 3.649434E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.878 | TFLOPs: 30.31 | 7: iteration 7030/ 21553 | consumed samples: 1799680 | consumed tokens: 3685744640 | elapsed time per iteration (s): 0.30 | learning rate: 1.584E-04 | global batch size: 256 | lm loss: 3.652820E+00 | grad norm: 0.379 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.963 | TFLOPs: 30.31 | 7: iteration 7040/ 21553 | consumed samples: 1802240 | consumed tokens: 3690987520 | elapsed time per iteration (s): 0.30 | learning rate: 1.583E-04 | global batch size: 256 | lm loss: 3.661848E+00 | grad norm: 0.477 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.112 | TFLOPs: 30.32 | 7: iteration 7050/ 21553 | consumed samples: 1804800 | consumed tokens: 3696230400 | elapsed time per iteration (s): 0.30 | learning rate: 1.582E-04 | global batch size: 256 | lm loss: 3.658066E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.653 | TFLOPs: 30.30 | 7: iteration 7060/ 21553 | consumed samples: 1807360 | consumed tokens: 3701473280 | elapsed time per iteration (s): 0.30 | learning rate: 1.580E-04 | global batch size: 256 | lm loss: 3.652993E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.828 | TFLOPs: 30.31 | 7: iteration 7070/ 21553 | consumed samples: 1809920 | consumed tokens: 3706716160 | elapsed time per iteration (s): 0.30 | learning rate: 1.579E-04 | global batch size: 256 | lm loss: 3.649855E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.618 | TFLOPs: 30.30 | 7: iteration 7080/ 21553 | consumed samples: 1812480 | consumed tokens: 3711959040 | elapsed time per iteration (s): 0.30 | learning rate: 1.578E-04 | global batch size: 256 | lm loss: 3.659213E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.171 | TFLOPs: 30.32 | 7: iteration 7090/ 21553 | consumed samples: 1815040 | consumed tokens: 3717201920 | elapsed time per iteration (s): 0.30 | learning rate: 1.577E-04 | global batch size: 256 | lm loss: 3.648584E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.898 | TFLOPs: 30.31 | 7: iteration 7100/ 21553 | consumed samples: 1817600 | consumed tokens: 3722444800 | elapsed time per iteration (s): 0.30 | learning rate: 1.576E-04 | global batch size: 256 | lm loss: 3.652576E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.655 | TFLOPs: 30.30 | 7: iteration 7110/ 21553 | consumed samples: 1820160 | consumed tokens: 3727687680 | elapsed time per iteration (s): 0.30 | learning rate: 1.575E-04 | global batch size: 256 | lm loss: 3.650577E+00 | grad norm: 0.374 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.985 | TFLOPs: 30.32 | 7: iteration 7120/ 21553 | consumed samples: 1822720 | consumed tokens: 3732930560 | elapsed time per iteration (s): 0.30 | learning rate: 1.574E-04 | global batch size: 256 | lm loss: 3.649532E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.224 | TFLOPs: 30.32 | 7: iteration 7130/ 21553 | consumed samples: 1825280 | consumed tokens: 3738173440 | elapsed time per iteration (s): 0.30 | learning rate: 1.573E-04 | global batch size: 256 | lm loss: 3.653909E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.935 | TFLOPs: 30.31 | 7: iteration 7140/ 21553 | consumed samples: 1827840 | consumed tokens: 3743416320 | elapsed time per iteration (s): 0.30 | learning rate: 1.571E-04 | global batch size: 256 | lm loss: 3.650692E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.842 | TFLOPs: 30.35 | 7: iteration 7150/ 21553 | consumed samples: 1830400 | consumed tokens: 3748659200 | elapsed time per iteration (s): 0.30 | learning rate: 1.570E-04 | global batch size: 256 | lm loss: 3.650932E+00 | grad norm: 0.370 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.118 | TFLOPs: 30.32 | 7: iteration 7160/ 21553 | consumed samples: 1832960 | consumed tokens: 3753902080 | elapsed time per iteration (s): 0.30 | learning rate: 1.569E-04 | global batch size: 256 | lm loss: 3.645731E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.443 | TFLOPs: 30.33 | 7: iteration 7170/ 21553 | consumed samples: 1835520 | consumed tokens: 3759144960 | elapsed time per iteration (s): 0.30 | learning rate: 1.568E-04 | global batch size: 256 | lm loss: 3.648368E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.691 | TFLOPs: 30.34 | 7: iteration 7180/ 21553 | consumed samples: 1838080 | consumed tokens: 3764387840 | elapsed time per iteration (s): 0.30 | learning rate: 1.567E-04 | global batch size: 256 | lm loss: 3.647275E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.292 | TFLOPs: 30.33 | 7: iteration 7190/ 21553 | consumed samples: 1840640 | consumed tokens: 3769630720 | elapsed time per iteration (s): 0.30 | learning rate: 1.566E-04 | global batch size: 256 | lm loss: 3.640462E+00 | grad norm: 0.383 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.308 | TFLOPs: 30.33 | 7: iteration 7200/ 21553 | consumed samples: 1843200 | consumed tokens: 3774873600 | elapsed time per iteration (s): 0.30 | learning rate: 1.565E-04 | global batch size: 256 | lm loss: 3.649248E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.536 | TFLOPs: 30.34 | 7: iteration 7210/ 21553 | consumed samples: 1845760 | consumed tokens: 3780116480 | elapsed time per iteration (s): 0.30 | learning rate: 1.563E-04 | global batch size: 256 | lm loss: 3.642511E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.148 | TFLOPs: 30.32 | 7: iteration 7220/ 21553 | consumed samples: 1848320 | consumed tokens: 3785359360 | elapsed time per iteration (s): 0.30 | learning rate: 1.562E-04 | global batch size: 256 | lm loss: 3.644078E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.336 | TFLOPs: 30.33 | 7: iteration 7230/ 21553 | consumed samples: 1850880 | consumed tokens: 3790602240 | elapsed time per iteration (s): 0.30 | learning rate: 1.561E-04 | global batch size: 256 | lm loss: 3.648815E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.160 | TFLOPs: 30.32 | 7: iteration 7240/ 21553 | consumed samples: 1853440 | consumed tokens: 3795845120 | elapsed time per iteration (s): 0.30 | learning rate: 1.560E-04 | global batch size: 256 | lm loss: 3.638037E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.354 | TFLOPs: 30.33 | 7: iteration 7250/ 21553 | consumed samples: 1856000 | consumed tokens: 3801088000 | elapsed time per iteration (s): 0.30 | learning rate: 1.559E-04 | global batch size: 256 | lm loss: 3.640165E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.966 | TFLOPs: 30.32 | 7: iteration 7260/ 21553 | consumed samples: 1858560 | consumed tokens: 3806330880 | elapsed time per iteration (s): 0.30 | learning rate: 1.558E-04 | global batch size: 256 | lm loss: 3.642400E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.202 | TFLOPs: 30.32 | 7: iteration 7270/ 21553 | consumed samples: 1861120 | consumed tokens: 3811573760 | elapsed time per iteration (s): 0.30 | learning rate: 1.557E-04 | global batch size: 256 | lm loss: 3.638821E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.248 | TFLOPs: 30.32 | 7: iteration 7280/ 21553 | consumed samples: 1863680 | consumed tokens: 3816816640 | elapsed time per iteration (s): 0.30 | learning rate: 1.556E-04 | global batch size: 256 | lm loss: 3.639407E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.344 | TFLOPs: 30.33 | 7: iteration 7290/ 21553 | consumed samples: 1866240 | consumed tokens: 3822059520 | elapsed time per iteration (s): 0.30 | learning rate: 1.554E-04 | global batch size: 256 | lm loss: 3.633954E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.521 | TFLOPs: 30.30 | 7: iteration 7300/ 21553 | consumed samples: 1868800 | consumed tokens: 3827302400 | elapsed time per iteration (s): 0.30 | learning rate: 1.553E-04 | global batch size: 256 | lm loss: 3.644241E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.776 | TFLOPs: 30.27 | 7: iteration 7310/ 21553 | consumed samples: 1871360 | consumed tokens: 3832545280 | elapsed time per iteration (s): 0.30 | learning rate: 1.552E-04 | global batch size: 256 | lm loss: 3.634974E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.968 | TFLOPs: 30.32 | 7: iteration 7320/ 21553 | consumed samples: 1873920 | consumed tokens: 3837788160 | elapsed time per iteration (s): 0.30 | learning rate: 1.551E-04 | global batch size: 256 | lm loss: 3.640311E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.267 | TFLOPs: 30.33 | 7: iteration 7330/ 21553 | consumed samples: 1876480 | consumed tokens: 3843031040 | elapsed time per iteration (s): 0.30 | learning rate: 1.550E-04 | global batch size: 256 | lm loss: 3.634700E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.312 | TFLOPs: 30.29 | 7: iteration 7340/ 21553 | consumed samples: 1879040 | consumed tokens: 3848273920 | elapsed time per iteration (s): 0.30 | learning rate: 1.549E-04 | global batch size: 256 | lm loss: 3.636813E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.608 | TFLOPs: 30.30 | 7: iteration 7350/ 21553 | consumed samples: 1881600 | consumed tokens: 3853516800 | elapsed time per iteration (s): 0.30 | learning rate: 1.547E-04 | global batch size: 256 | lm loss: 3.633150E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.547 | TFLOPs: 30.30 | 7: iteration 7360/ 21553 | consumed samples: 1884160 | consumed tokens: 3858759680 | elapsed time per iteration (s): 0.30 | learning rate: 1.546E-04 | global batch size: 256 | lm loss: 3.629980E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.750 | TFLOPs: 30.31 | 7: iteration 7370/ 21553 | consumed samples: 1886720 | consumed tokens: 3864002560 | elapsed time per iteration (s): 0.30 | learning rate: 1.545E-04 | global batch size: 256 | lm loss: 3.634319E+00 | grad norm: 0.384 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.556 | TFLOPs: 30.30 | 7: iteration 7380/ 21553 | consumed samples: 1889280 | consumed tokens: 3869245440 | elapsed time per iteration (s): 0.30 | learning rate: 1.544E-04 | global batch size: 256 | lm loss: 3.635736E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.955 | TFLOPs: 30.31 | 7: iteration 7390/ 21553 | consumed samples: 1891840 | consumed tokens: 3874488320 | elapsed time per iteration (s): 0.30 | learning rate: 1.543E-04 | global batch size: 256 | lm loss: 3.639292E+00 | grad norm: 0.464 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.792 | TFLOPs: 30.31 | 7: iteration 7400/ 21553 | consumed samples: 1894400 | consumed tokens: 3879731200 | elapsed time per iteration (s): 0.30 | learning rate: 1.542E-04 | global batch size: 256 | lm loss: 3.632040E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.706 | TFLOPs: 30.31 | 7: iteration 7410/ 21553 | consumed samples: 1896960 | consumed tokens: 3884974080 | elapsed time per iteration (s): 0.30 | learning rate: 1.541E-04 | global batch size: 256 | lm loss: 3.623086E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.282 | TFLOPs: 30.33 | 7: iteration 7420/ 21553 | consumed samples: 1899520 | consumed tokens: 3890216960 | elapsed time per iteration (s): 0.30 | learning rate: 1.539E-04 | global batch size: 256 | lm loss: 3.628849E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.819 | TFLOPs: 30.31 | 7: iteration 7430/ 21553 | consumed samples: 1902080 | consumed tokens: 3895459840 | elapsed time per iteration (s): 0.30 | learning rate: 1.538E-04 | global batch size: 256 | lm loss: 3.636912E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.724 | TFLOPs: 30.31 | 7: iteration 7440/ 21553 | consumed samples: 1904640 | consumed tokens: 3900702720 | elapsed time per iteration (s): 0.30 | learning rate: 1.537E-04 | global batch size: 256 | lm loss: 3.620575E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.834 | TFLOPs: 30.31 | 7: iteration 7450/ 21553 | consumed samples: 1907200 | consumed tokens: 3905945600 | elapsed time per iteration (s): 0.30 | learning rate: 1.536E-04 | global batch size: 256 | lm loss: 3.630096E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.572 | TFLOPs: 30.30 | 7: iteration 7460/ 21553 | consumed samples: 1909760 | consumed tokens: 3911188480 | elapsed time per iteration (s): 0.30 | learning rate: 1.535E-04 | global batch size: 256 | lm loss: 3.632698E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.800 | TFLOPs: 30.31 | 7: iteration 7470/ 21553 | consumed samples: 1912320 | consumed tokens: 3916431360 | elapsed time per iteration (s): 0.30 | learning rate: 1.534E-04 | global batch size: 256 | lm loss: 3.633292E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.686 | TFLOPs: 30.31 | 7: iteration 7480/ 21553 | consumed samples: 1914880 | consumed tokens: 3921674240 | elapsed time per iteration (s): 0.30 | learning rate: 1.532E-04 | global batch size: 256 | lm loss: 3.632732E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.997 | TFLOPs: 30.32 | 7: iteration 7490/ 21553 | consumed samples: 1917440 | consumed tokens: 3926917120 | elapsed time per iteration (s): 0.30 | learning rate: 1.531E-04 | global batch size: 256 | lm loss: 3.622918E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.933 | TFLOPs: 30.31 | 7: iteration 7500/ 21553 | consumed samples: 1920000 | consumed tokens: 3932160000 | elapsed time per iteration (s): 0.30 | learning rate: 1.530E-04 | global batch size: 256 | lm loss: 3.631092E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.775 | TFLOPs: 30.31 | 7: iteration 7510/ 21553 | consumed samples: 1922560 | consumed tokens: 3937402880 | elapsed time per iteration (s): 0.30 | learning rate: 1.529E-04 | global batch size: 256 | lm loss: 3.628968E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.053 | TFLOPs: 30.32 | 7: iteration 7520/ 21553 | consumed samples: 1925120 | consumed tokens: 3942645760 | elapsed time per iteration (s): 0.30 | learning rate: 1.528E-04 | global batch size: 256 | lm loss: 3.622496E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.682 | TFLOPs: 30.31 | 7: iteration 7530/ 21553 | consumed samples: 1927680 | consumed tokens: 3947888640 | elapsed time per iteration (s): 0.30 | learning rate: 1.527E-04 | global batch size: 256 | lm loss: 3.626979E+00 | grad norm: 0.385 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.450 | TFLOPs: 30.30 | 7: iteration 7540/ 21553 | consumed samples: 1930240 | consumed tokens: 3953131520 | elapsed time per iteration (s): 0.30 | learning rate: 1.525E-04 | global batch size: 256 | lm loss: 3.628159E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.166 | TFLOPs: 30.29 | 7: iteration 7550/ 21553 | consumed samples: 1932800 | consumed tokens: 3958374400 | elapsed time per iteration (s): 0.30 | learning rate: 1.524E-04 | global batch size: 256 | lm loss: 3.626472E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.434 | TFLOPs: 30.30 | 7: iteration 7560/ 21553 | consumed samples: 1935360 | consumed tokens: 3963617280 | elapsed time per iteration (s): 0.30 | learning rate: 1.523E-04 | global batch size: 256 | lm loss: 3.628358E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.660 | TFLOPs: 30.30 | 7: iteration 7570/ 21553 | consumed samples: 1937920 | consumed tokens: 3968860160 | elapsed time per iteration (s): 0.30 | learning rate: 1.522E-04 | global batch size: 256 | lm loss: 3.620074E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.409 | TFLOPs: 30.30 | 7: iteration 7580/ 21553 | consumed samples: 1940480 | consumed tokens: 3974103040 | elapsed time per iteration (s): 0.30 | learning rate: 1.521E-04 | global batch size: 256 | lm loss: 3.625449E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.311 | TFLOPs: 30.29 | 7: iteration 7590/ 21553 | consumed samples: 1943040 | consumed tokens: 3979345920 | elapsed time per iteration (s): 0.30 | learning rate: 1.520E-04 | global batch size: 256 | lm loss: 3.622480E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.458 | TFLOPs: 30.30 | 7: iteration 7600/ 21553 | consumed samples: 1945600 | consumed tokens: 3984588800 | elapsed time per iteration (s): 0.30 | learning rate: 1.518E-04 | global batch size: 256 | lm loss: 3.632582E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.541 | TFLOPs: 30.30 | 7: iteration 7610/ 21553 | consumed samples: 1948160 | consumed tokens: 3989831680 | elapsed time per iteration (s): 0.30 | learning rate: 1.517E-04 | global batch size: 256 | lm loss: 3.621811E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.750 | TFLOPs: 30.31 | 7: iteration 7620/ 21553 | consumed samples: 1950720 | consumed tokens: 3995074560 | elapsed time per iteration (s): 0.30 | learning rate: 1.516E-04 | global batch size: 256 | lm loss: 3.619495E+00 | grad norm: 0.384 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.893 | TFLOPs: 30.31 | 7: iteration 7630/ 21553 | consumed samples: 1953280 | consumed tokens: 4000317440 | elapsed time per iteration (s): 0.30 | learning rate: 1.515E-04 | global batch size: 256 | lm loss: 3.614981E+00 | grad norm: 0.378 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.650 | TFLOPs: 30.30 | 7: iteration 7640/ 21553 | consumed samples: 1955840 | consumed tokens: 4005560320 | elapsed time per iteration (s): 0.30 | learning rate: 1.514E-04 | global batch size: 256 | lm loss: 3.622057E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.680 | TFLOPs: 30.31 | 7: iteration 7650/ 21553 | consumed samples: 1958400 | consumed tokens: 4010803200 | elapsed time per iteration (s): 0.30 | learning rate: 1.513E-04 | global batch size: 256 | lm loss: 3.625505E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.629 | TFLOPs: 30.30 | 7: iteration 7660/ 21553 | consumed samples: 1960960 | consumed tokens: 4016046080 | elapsed time per iteration (s): 0.30 | learning rate: 1.511E-04 | global batch size: 256 | lm loss: 3.619876E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.180 | TFLOPs: 30.29 | 7: iteration 7670/ 21553 | consumed samples: 1963520 | consumed tokens: 4021288960 | elapsed time per iteration (s): 0.30 | learning rate: 1.510E-04 | global batch size: 256 | lm loss: 3.616962E+00 | grad norm: 0.458 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.958 | TFLOPs: 30.28 | 7: iteration 7680/ 21553 | consumed samples: 1966080 | consumed tokens: 4026531840 | elapsed time per iteration (s): 0.30 | learning rate: 1.509E-04 | global batch size: 256 | lm loss: 3.618737E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.214 | TFLOPs: 30.29 | 7: iteration 7690/ 21553 | consumed samples: 1968640 | consumed tokens: 4031774720 | elapsed time per iteration (s): 0.30 | learning rate: 1.508E-04 | global batch size: 256 | lm loss: 3.623339E+00 | grad norm: 0.382 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.392 | TFLOPs: 30.29 | 7: iteration 7700/ 21553 | consumed samples: 1971200 | consumed tokens: 4037017600 | elapsed time per iteration (s): 0.30 | learning rate: 1.507E-04 | global batch size: 256 | lm loss: 3.616772E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.375 | TFLOPs: 30.29 | 7: iteration 7710/ 21553 | consumed samples: 1973760 | consumed tokens: 4042260480 | elapsed time per iteration (s): 0.30 | learning rate: 1.505E-04 | global batch size: 256 | lm loss: 3.622328E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.380 | TFLOPs: 30.29 | 7: iteration 7720/ 21553 | consumed samples: 1976320 | consumed tokens: 4047503360 | elapsed time per iteration (s): 0.30 | learning rate: 1.504E-04 | global batch size: 256 | lm loss: 3.627041E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.395 | TFLOPs: 30.30 | 7: iteration 7730/ 21553 | consumed samples: 1978880 | consumed tokens: 4052746240 | elapsed time per iteration (s): 0.30 | learning rate: 1.503E-04 | global batch size: 256 | lm loss: 3.614056E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.106 | TFLOPs: 30.28 | 7: iteration 7740/ 21553 | consumed samples: 1981440 | consumed tokens: 4057989120 | elapsed time per iteration (s): 0.30 | learning rate: 1.502E-04 | global batch size: 256 | lm loss: 3.619865E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.915 | TFLOPs: 30.31 | 7: iteration 7750/ 21553 | consumed samples: 1984000 | consumed tokens: 4063232000 | elapsed time per iteration (s): 0.30 | learning rate: 1.501E-04 | global batch size: 256 | lm loss: 3.619549E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.546 | TFLOPs: 30.30 | 7: iteration 7760/ 21553 | consumed samples: 1986560 | consumed tokens: 4068474880 | elapsed time per iteration (s): 0.30 | learning rate: 1.500E-04 | global batch size: 256 | lm loss: 3.612350E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.201 | TFLOPs: 30.29 | 7: iteration 7770/ 21553 | consumed samples: 1989120 | consumed tokens: 4073717760 | elapsed time per iteration (s): 0.30 | learning rate: 1.498E-04 | global batch size: 256 | lm loss: 3.613172E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.619 | TFLOPs: 30.30 | 7: iteration 7780/ 21553 | consumed samples: 1991680 | consumed tokens: 4078960640 | elapsed time per iteration (s): 0.30 | learning rate: 1.497E-04 | global batch size: 256 | lm loss: 3.616495E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.326 | TFLOPs: 30.29 | 7: iteration 7790/ 21553 | consumed samples: 1994240 | consumed tokens: 4084203520 | elapsed time per iteration (s): 0.30 | learning rate: 1.496E-04 | global batch size: 256 | lm loss: 3.622307E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.829 | TFLOPs: 30.31 | 7: iteration 7800/ 21553 | consumed samples: 1996800 | consumed tokens: 4089446400 | elapsed time per iteration (s): 0.30 | learning rate: 1.495E-04 | global batch size: 256 | lm loss: 3.612967E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.869 | TFLOPs: 30.31 | 7: iteration 7810/ 21553 | consumed samples: 1999360 | consumed tokens: 4094689280 | elapsed time per iteration (s): 0.30 | learning rate: 1.494E-04 | global batch size: 256 | lm loss: 3.618262E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.172 | TFLOPs: 30.29 | 7: iteration 7820/ 21553 | consumed samples: 2001920 | consumed tokens: 4099932160 | elapsed time per iteration (s): 0.30 | learning rate: 1.492E-04 | global batch size: 256 | lm loss: 3.609213E+00 | grad norm: 0.385 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.170 | TFLOPs: 30.29 | 7: iteration 7830/ 21553 | consumed samples: 2004480 | consumed tokens: 4105175040 | elapsed time per iteration (s): 0.30 | learning rate: 1.491E-04 | global batch size: 256 | lm loss: 3.605101E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.179 | TFLOPs: 30.29 | 7: iteration 7840/ 21553 | consumed samples: 2007040 | consumed tokens: 4110417920 | elapsed time per iteration (s): 0.30 | learning rate: 1.490E-04 | global batch size: 256 | lm loss: 3.615295E+00 | grad norm: 0.451 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.800 | TFLOPs: 30.31 | 7: iteration 7850/ 21553 | consumed samples: 2009600 | consumed tokens: 4115660800 | elapsed time per iteration (s): 0.30 | learning rate: 1.489E-04 | global batch size: 256 | lm loss: 3.610882E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.723 | TFLOPs: 30.31 | 7: iteration 7860/ 21553 | consumed samples: 2012160 | consumed tokens: 4120903680 | elapsed time per iteration (s): 0.30 | learning rate: 1.488E-04 | global batch size: 256 | lm loss: 3.609616E+00 | grad norm: 0.373 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.762 | TFLOPs: 30.31 | 7: iteration 7870/ 21553 | consumed samples: 2014720 | consumed tokens: 4126146560 | elapsed time per iteration (s): 0.30 | learning rate: 1.486E-04 | global batch size: 256 | lm loss: 3.606347E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.780 | TFLOPs: 30.31 | 7: iteration 7880/ 21553 | consumed samples: 2017280 | consumed tokens: 4131389440 | elapsed time per iteration (s): 0.30 | learning rate: 1.485E-04 | global batch size: 256 | lm loss: 3.605713E+00 | grad norm: 0.383 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.322 | TFLOPs: 30.29 | 7: iteration 7890/ 21553 | consumed samples: 2019840 | consumed tokens: 4136632320 | elapsed time per iteration (s): 0.30 | learning rate: 1.484E-04 | global batch size: 256 | lm loss: 3.615512E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.688 | TFLOPs: 30.31 | 7: iteration 7900/ 21553 | consumed samples: 2022400 | consumed tokens: 4141875200 | elapsed time per iteration (s): 0.30 | learning rate: 1.483E-04 | global batch size: 256 | lm loss: 3.606553E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.185 | TFLOPs: 30.29 | 7: iteration 7910/ 21553 | consumed samples: 2024960 | consumed tokens: 4147118080 | elapsed time per iteration (s): 0.30 | learning rate: 1.482E-04 | global batch size: 256 | lm loss: 3.608429E+00 | grad norm: 0.379 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.284 | TFLOPs: 30.29 | 7: iteration 7920/ 21553 | consumed samples: 2027520 | consumed tokens: 4152360960 | elapsed time per iteration (s): 0.30 | learning rate: 1.480E-04 | global batch size: 256 | lm loss: 3.609279E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.700 | TFLOPs: 30.31 | 7: iteration 7930/ 21553 | consumed samples: 2030080 | consumed tokens: 4157603840 | elapsed time per iteration (s): 0.30 | learning rate: 1.479E-04 | global batch size: 256 | lm loss: 3.607884E+00 | grad norm: 0.397 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.437 | TFLOPs: 30.30 | 7: iteration 7940/ 21553 | consumed samples: 2032640 | consumed tokens: 4162846720 | elapsed time per iteration (s): 0.30 | learning rate: 1.478E-04 | global batch size: 256 | lm loss: 3.609647E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.174 | TFLOPs: 30.29 | 7: iteration 7950/ 21553 | consumed samples: 2035200 | consumed tokens: 4168089600 | elapsed time per iteration (s): 0.30 | learning rate: 1.477E-04 | global batch size: 256 | lm loss: 3.606245E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.882 | TFLOPs: 30.28 | 7: iteration 7960/ 21553 | consumed samples: 2037760 | consumed tokens: 4173332480 | elapsed time per iteration (s): 0.30 | learning rate: 1.476E-04 | global batch size: 256 | lm loss: 3.610137E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.140 | TFLOPs: 30.29 | 7: iteration 7970/ 21553 | consumed samples: 2040320 | consumed tokens: 4178575360 | elapsed time per iteration (s): 0.30 | learning rate: 1.474E-04 | global batch size: 256 | lm loss: 3.603700E+00 | grad norm: 0.380 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.265 | TFLOPs: 30.29 | 7: iteration 7980/ 21553 | consumed samples: 2042880 | consumed tokens: 4183818240 | elapsed time per iteration (s): 0.30 | learning rate: 1.473E-04 | global batch size: 256 | lm loss: 3.597428E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.814 | TFLOPs: 30.27 | 7: iteration 7990/ 21553 | consumed samples: 2045440 | consumed tokens: 4189061120 | elapsed time per iteration (s): 0.30 | learning rate: 1.472E-04 | global batch size: 256 | lm loss: 3.603948E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.160 | TFLOPs: 30.29 | 0: [2023-03-17 00:35:33,178] [INFO] [logging.py:68:log_dist] [Rank 0] step=8000, skipped=0, lr=[0.00014708123251590496, 0.00014708123251590496, 0.00014708123251590496], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 8000/ 21553 | consumed samples: 2048000 | consumed tokens: 4194304000 | elapsed time per iteration (s): 0.30 | learning rate: 1.471E-04 | global batch size: 256 | lm loss: 3.599217E+00 | grad norm: 0.377 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.020 | TFLOPs: 30.28 | 0: steps: 8000 loss: 3.5793 iter time (s): 0.294 samples/sec: 871.766 7: ----------------------------------------------------------------------------------------------- 7: validation loss at iteration 8000 | lm loss value: 3.919360E+00 | lm loss PPL: 5.036821E+01 | 7: ----------------------------------------------------------------------------------------------- 0: saving checkpoint at iteration 8000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:35:33,299] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step8000 is begin to save! 0: [2023-03-17 00:35:33,302] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:35:33,387] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:35:33,387] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:35:33,403] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:35:33,404] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:35:33,418] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:35:33,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:35:33,433] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:35:33,433] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:35:33,448] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:35:33,448] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:35:33,463] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:35:33,463] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:35:33,477] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:35:33,478] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:35:33,492] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:35:33,492] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:35:33,507] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:35:33,507] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:35:33,522] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:35:33,522] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:35:33,536] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:35:33,537] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:35:33,551] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:35:33,551] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:35:33,566] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:35:33,566] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:35:33,581] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:35:33,581] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:35:33,596] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:35:33,596] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:35:33,611] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:35:33,611] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:35:33,612] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:35:33,613] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step8000/mp_rank_00_model_states.pt 0: [2023-03-17 00:35:33,613] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:35:33,615] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:35:33,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:35:33,663] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:35:33,663] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:35:33,663] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 6: [2023-03-17 00:35:33,665] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:35:33,665] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:35:33,665] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 0: [2023-03-17 00:35:33,666] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:35:33,666] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:35:33,666] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 00:35:33,666] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 0: [2023-03-17 00:35:33,667] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:35:33,667] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:35:33,667] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 1: [2023-03-17 00:35:33,667] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:35:33,667] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:35:33,667] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 6: [2023-03-17 00:35:33,667] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:35:33,667] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:35:33,667] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 2: [2023-03-17 00:35:33,667] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:35:33,668] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 0: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:35:33,668] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 3: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:35:33,668] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 00:35:33,668] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:35:33,668] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 3: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 3: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 3: [2023-03-17 00:35:33,669] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 1: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 1: [2023-03-17 00:35:33,669] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 2: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:35:33,669] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 1: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:35:33,669] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 00:35:33,669] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 1: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 0: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:35:33,669] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 0: [2023-03-17 00:35:33,669] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 2: [2023-03-17 00:35:33,669] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 6: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:35:33,670] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 00:35:33,670] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 6: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 4: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:35:33,670] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 3: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:35:33,670] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 2: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:35:33,671] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:35:33,671] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 2: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 5: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:35:33,671] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 5: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:35:33,671] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 4: [2023-03-17 00:35:33,670] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:35:33,670] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 4: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 5: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:35:33,671] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 5: [2023-03-17 00:35:33,671] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 4: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:35:33,671] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 4: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:35:33,672] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 00:35:33,672] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 3: [2023-03-17 00:35:33,672] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:35:33,672] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 00:35:33,672] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 5: [2023-03-17 00:35:33,672] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:35:33,672] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 00:35:33,672] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 5: [2023-03-17 00:35:33,672] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:35:33,672] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:35:33,672] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 6: [2023-03-17 00:35:33,673] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:35:33,673] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:35:33,673] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 2: [2023-03-17 00:35:33,673] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 6: [2023-03-17 00:35:33,673] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 2: [2023-03-17 00:35:33,673] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 4: [2023-03-17 00:35:33,673] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:35:33,674] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:35:33,674] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:35:33,674] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 2: [2023-03-17 00:35:33,674] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:35:33,674] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:35:33,674] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 1: [2023-03-17 00:35:33,675] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:35:33,675] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:35:33,675] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 5: [2023-03-17 00:35:33,675] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:35:33,675] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 00:35:33,675] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 1: [2023-03-17 00:35:33,675] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:35:33,675] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:35:33,675] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:35:33,675] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 3: [2023-03-17 00:35:33,675] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 5: [2023-03-17 00:35:33,676] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 1: [2023-03-17 00:35:33,676] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 3: [2023-03-17 00:35:33,676] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 5: [2023-03-17 00:35:33,676] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 6: [2023-03-17 00:35:33,676] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:35:33,676] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:35:33,676] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 6: [2023-03-17 00:35:33,676] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:35:33,676] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 00:35:33,676] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,664] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:35:33,677] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:35:33,677] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:35:33,677] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 00:35:33,677] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 5: [2023-03-17 00:35:33,677] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:35:33,677] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 0: [2023-03-17 00:35:33,677] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:35:33,678] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 00:35:33,678] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 3: [2023-03-17 00:35:33,678] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:35:33,678] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 00:35:33,678] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 4: [2023-03-17 00:35:33,674] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 00:35:33,674] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 4: [2023-03-17 00:35:33,674] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:35:33,674] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 00:35:33,674] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 4: [2023-03-17 00:35:33,675] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:35:33,675] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 00:35:33,675] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 2: [2023-03-17 00:35:33,678] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:35:33,679] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:35:33,679] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 7: [2023-03-17 00:35:33,664] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 00:35:33,664] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,667] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:35:33,679] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 2: [2023-03-17 00:35:33,679] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,667] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:35:33,668] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:35:33,670] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 0: [2023-03-17 00:35:33,679] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:35:33,670] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:35:33,670] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:35:33,671] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 00:35:33,671] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,672] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:35:33,672] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 00:35:33,672] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,673] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:35:33,673] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 00:35:33,674] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 7: [2023-03-17 00:35:33,677] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:35:33,677] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 00:35:33,677] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 0: [2023-03-17 00:35:33,685] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step8000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:35:33,685] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step8000 is ready now! 0: successfully saved checkpoint at iteration 8000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 396.07 7: iteration 8010/ 21553 | consumed samples: 2050560 | consumed tokens: 4199546880 | elapsed time per iteration (s): 0.35 | learning rate: 1.470E-04 | global batch size: 256 | lm loss: 3.594522E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 731.126 | TFLOPs: 25.59 | 7: iteration 8020/ 21553 | consumed samples: 2053120 | consumed tokens: 4204789760 | elapsed time per iteration (s): 0.30 | learning rate: 1.468E-04 | global batch size: 256 | lm loss: 3.602663E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.518 | TFLOPs: 30.30 | 7: iteration 8030/ 21553 | consumed samples: 2055680 | consumed tokens: 4210032640 | elapsed time per iteration (s): 0.30 | learning rate: 1.467E-04 | global batch size: 256 | lm loss: 3.606510E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.952 | TFLOPs: 30.31 | 7: iteration 8040/ 21553 | consumed samples: 2058240 | consumed tokens: 4215275520 | elapsed time per iteration (s): 0.30 | learning rate: 1.466E-04 | global batch size: 256 | lm loss: 3.606675E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.241 | TFLOPs: 30.32 | 7: iteration 8050/ 21553 | consumed samples: 2060800 | consumed tokens: 4220518400 | elapsed time per iteration (s): 0.30 | learning rate: 1.465E-04 | global batch size: 256 | lm loss: 3.601995E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.809 | TFLOPs: 30.31 | 7: iteration 8060/ 21553 | consumed samples: 2063360 | consumed tokens: 4225761280 | elapsed time per iteration (s): 0.30 | learning rate: 1.464E-04 | global batch size: 256 | lm loss: 3.605812E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.207 | TFLOPs: 30.29 | 7: iteration 8070/ 21553 | consumed samples: 2065920 | consumed tokens: 4231004160 | elapsed time per iteration (s): 0.30 | learning rate: 1.462E-04 | global batch size: 256 | lm loss: 3.600177E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.299 | TFLOPs: 30.29 | 7: iteration 8080/ 21553 | consumed samples: 2068480 | consumed tokens: 4236247040 | elapsed time per iteration (s): 0.30 | learning rate: 1.461E-04 | global batch size: 256 | lm loss: 3.598959E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.639 | TFLOPs: 30.27 | 7: iteration 8090/ 21553 | consumed samples: 2071040 | consumed tokens: 4241489920 | elapsed time per iteration (s): 0.30 | learning rate: 1.460E-04 | global batch size: 256 | lm loss: 3.598211E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.743 | TFLOPs: 30.27 | 7: iteration 8100/ 21553 | consumed samples: 2073600 | consumed tokens: 4246732800 | elapsed time per iteration (s): 0.30 | learning rate: 1.459E-04 | global batch size: 256 | lm loss: 3.600976E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.650 | TFLOPs: 30.23 | 7: iteration 8110/ 21553 | consumed samples: 2076160 | consumed tokens: 4251975680 | elapsed time per iteration (s): 0.30 | learning rate: 1.457E-04 | global batch size: 256 | lm loss: 3.597301E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 860.064 | TFLOPs: 30.11 | 7: iteration 8120/ 21553 | consumed samples: 2078720 | consumed tokens: 4257218560 | elapsed time per iteration (s): 0.30 | learning rate: 1.456E-04 | global batch size: 256 | lm loss: 3.597935E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.191 | TFLOPs: 30.15 | 7: iteration 8130/ 21553 | consumed samples: 2081280 | consumed tokens: 4262461440 | elapsed time per iteration (s): 0.30 | learning rate: 1.455E-04 | global batch size: 256 | lm loss: 3.596959E+00 | grad norm: 0.388 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.456 | TFLOPs: 30.16 | 7: iteration 8140/ 21553 | consumed samples: 2083840 | consumed tokens: 4267704320 | elapsed time per iteration (s): 0.30 | learning rate: 1.454E-04 | global batch size: 256 | lm loss: 3.602731E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.027 | TFLOPs: 30.25 | 7: iteration 8150/ 21553 | consumed samples: 2086400 | consumed tokens: 4272947200 | elapsed time per iteration (s): 0.30 | learning rate: 1.453E-04 | global batch size: 256 | lm loss: 3.599988E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.359 | TFLOPs: 30.26 | 7: iteration 8160/ 21553 | consumed samples: 2088960 | consumed tokens: 4278190080 | elapsed time per iteration (s): 0.30 | learning rate: 1.451E-04 | global batch size: 256 | lm loss: 3.600715E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.657 | TFLOPs: 29.99 | 7: iteration 8170/ 21553 | consumed samples: 2091520 | consumed tokens: 4283432960 | elapsed time per iteration (s): 0.30 | learning rate: 1.450E-04 | global batch size: 256 | lm loss: 3.597445E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.824 | TFLOPs: 30.24 | 7: iteration 8180/ 21553 | consumed samples: 2094080 | consumed tokens: 4288675840 | elapsed time per iteration (s): 0.30 | learning rate: 1.449E-04 | global batch size: 256 | lm loss: 3.592639E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.097 | TFLOPs: 30.25 | 7: iteration 8190/ 21553 | consumed samples: 2096640 | consumed tokens: 4293918720 | elapsed time per iteration (s): 0.30 | learning rate: 1.448E-04 | global batch size: 256 | lm loss: 3.599971E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.982 | TFLOPs: 30.28 | 7: iteration 8200/ 21553 | consumed samples: 2099200 | consumed tokens: 4299161600 | elapsed time per iteration (s): 0.30 | learning rate: 1.447E-04 | global batch size: 256 | lm loss: 3.590911E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.835 | TFLOPs: 30.28 | 7: iteration 8210/ 21553 | consumed samples: 2101760 | consumed tokens: 4304404480 | elapsed time per iteration (s): 0.30 | learning rate: 1.445E-04 | global batch size: 256 | lm loss: 3.595802E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.419 | TFLOPs: 30.30 | 7: iteration 8220/ 21553 | consumed samples: 2104320 | consumed tokens: 4309647360 | elapsed time per iteration (s): 0.30 | learning rate: 1.444E-04 | global batch size: 256 | lm loss: 3.593998E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.719 | TFLOPs: 30.31 | 7: iteration 8230/ 21553 | consumed samples: 2106880 | consumed tokens: 4314890240 | elapsed time per iteration (s): 0.30 | learning rate: 1.443E-04 | global batch size: 256 | lm loss: 3.591462E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.547 | TFLOPs: 30.30 | 7: iteration 8240/ 21553 | consumed samples: 2109440 | consumed tokens: 4320133120 | elapsed time per iteration (s): 0.30 | learning rate: 1.442E-04 | global batch size: 256 | lm loss: 3.597606E+00 | grad norm: 0.379 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.293 | TFLOPs: 30.29 | 7: iteration 8250/ 21553 | consumed samples: 2112000 | consumed tokens: 4325376000 | elapsed time per iteration (s): 0.30 | learning rate: 1.440E-04 | global batch size: 256 | lm loss: 3.594201E+00 | grad norm: 0.367 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.796 | TFLOPs: 30.31 | 7: iteration 8260/ 21553 | consumed samples: 2114560 | consumed tokens: 4330618880 | elapsed time per iteration (s): 0.30 | learning rate: 1.439E-04 | global batch size: 256 | lm loss: 3.591533E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.825 | TFLOPs: 30.31 | 7: iteration 8270/ 21553 | consumed samples: 2117120 | consumed tokens: 4335861760 | elapsed time per iteration (s): 0.30 | learning rate: 1.438E-04 | global batch size: 256 | lm loss: 3.586121E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.384 | TFLOPs: 30.29 | 7: iteration 8280/ 21553 | consumed samples: 2119680 | consumed tokens: 4341104640 | elapsed time per iteration (s): 0.30 | learning rate: 1.437E-04 | global batch size: 256 | lm loss: 3.591221E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.851 | TFLOPs: 30.31 | 7: iteration 8290/ 21553 | consumed samples: 2122240 | consumed tokens: 4346347520 | elapsed time per iteration (s): 0.30 | learning rate: 1.435E-04 | global batch size: 256 | lm loss: 3.589232E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.486 | TFLOPs: 30.30 | 7: iteration 8300/ 21553 | consumed samples: 2124800 | consumed tokens: 4351590400 | elapsed time per iteration (s): 0.30 | learning rate: 1.434E-04 | global batch size: 256 | lm loss: 3.591024E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.301 | TFLOPs: 30.29 | 7: iteration 8310/ 21553 | consumed samples: 2127360 | consumed tokens: 4356833280 | elapsed time per iteration (s): 0.30 | learning rate: 1.433E-04 | global batch size: 256 | lm loss: 3.596430E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.154 | TFLOPs: 30.29 | 7: iteration 8320/ 21553 | consumed samples: 2129920 | consumed tokens: 4362076160 | elapsed time per iteration (s): 0.30 | learning rate: 1.432E-04 | global batch size: 256 | lm loss: 3.594726E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.833 | TFLOPs: 30.24 | 7: iteration 8330/ 21553 | consumed samples: 2132480 | consumed tokens: 4367319040 | elapsed time per iteration (s): 0.30 | learning rate: 1.431E-04 | global batch size: 256 | lm loss: 3.591145E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.758 | TFLOPs: 30.20 | 7: iteration 8340/ 21553 | consumed samples: 2135040 | consumed tokens: 4372561920 | elapsed time per iteration (s): 0.30 | learning rate: 1.429E-04 | global batch size: 256 | lm loss: 3.591186E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.352 | TFLOPs: 30.22 | 7: iteration 8350/ 21553 | consumed samples: 2137600 | consumed tokens: 4377804800 | elapsed time per iteration (s): 0.30 | learning rate: 1.428E-04 | global batch size: 256 | lm loss: 3.585690E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.881 | TFLOPs: 30.21 | 7: iteration 8360/ 21553 | consumed samples: 2140160 | consumed tokens: 4383047680 | elapsed time per iteration (s): 0.30 | learning rate: 1.427E-04 | global batch size: 256 | lm loss: 3.586066E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.068 | TFLOPs: 30.21 | 7: iteration 8370/ 21553 | consumed samples: 2142720 | consumed tokens: 4388290560 | elapsed time per iteration (s): 0.30 | learning rate: 1.426E-04 | global batch size: 256 | lm loss: 3.580708E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.849 | TFLOPs: 30.21 | 7: iteration 8380/ 21553 | consumed samples: 2145280 | consumed tokens: 4393533440 | elapsed time per iteration (s): 0.30 | learning rate: 1.424E-04 | global batch size: 256 | lm loss: 3.581649E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.565 | TFLOPs: 30.23 | 7: iteration 8390/ 21553 | consumed samples: 2147840 | consumed tokens: 4398776320 | elapsed time per iteration (s): 0.30 | learning rate: 1.423E-04 | global batch size: 256 | lm loss: 3.586951E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.326 | TFLOPs: 30.22 | 7: iteration 8400/ 21553 | consumed samples: 2150400 | consumed tokens: 4404019200 | elapsed time per iteration (s): 0.30 | learning rate: 1.422E-04 | global batch size: 256 | lm loss: 3.589030E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.154 | TFLOPs: 30.22 | 7: iteration 8410/ 21553 | consumed samples: 2152960 | consumed tokens: 4409262080 | elapsed time per iteration (s): 0.30 | learning rate: 1.421E-04 | global batch size: 256 | lm loss: 3.587185E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.396 | TFLOPs: 30.23 | 7: iteration 8420/ 21553 | consumed samples: 2155520 | consumed tokens: 4414504960 | elapsed time per iteration (s): 0.30 | learning rate: 1.419E-04 | global batch size: 256 | lm loss: 3.585764E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.005 | TFLOPs: 30.21 | 7: iteration 8430/ 21553 | consumed samples: 2158080 | consumed tokens: 4419747840 | elapsed time per iteration (s): 0.30 | learning rate: 1.418E-04 | global batch size: 256 | lm loss: 3.580065E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.985 | TFLOPs: 30.21 | 7: iteration 8440/ 21553 | consumed samples: 2160640 | consumed tokens: 4424990720 | elapsed time per iteration (s): 0.30 | learning rate: 1.417E-04 | global batch size: 256 | lm loss: 3.585640E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.339 | TFLOPs: 30.19 | 7: iteration 8450/ 21553 | consumed samples: 2163200 | consumed tokens: 4430233600 | elapsed time per iteration (s): 0.30 | learning rate: 1.416E-04 | global batch size: 256 | lm loss: 3.591322E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.262 | TFLOPs: 30.22 | 7: iteration 8460/ 21553 | consumed samples: 2165760 | consumed tokens: 4435476480 | elapsed time per iteration (s): 0.30 | learning rate: 1.414E-04 | global batch size: 256 | lm loss: 3.587804E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.759 | TFLOPs: 30.27 | 7: iteration 8470/ 21553 | consumed samples: 2168320 | consumed tokens: 4440719360 | elapsed time per iteration (s): 0.30 | learning rate: 1.413E-04 | global batch size: 256 | lm loss: 3.581151E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.718 | TFLOPs: 30.27 | 7: iteration 8480/ 21553 | consumed samples: 2170880 | consumed tokens: 4445962240 | elapsed time per iteration (s): 0.30 | learning rate: 1.412E-04 | global batch size: 256 | lm loss: 3.589733E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.067 | TFLOPs: 30.28 | 7: iteration 8490/ 21553 | consumed samples: 2173440 | consumed tokens: 4451205120 | elapsed time per iteration (s): 0.30 | learning rate: 1.411E-04 | global batch size: 256 | lm loss: 3.583441E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.358 | TFLOPs: 30.29 | 7: iteration 8500/ 21553 | consumed samples: 2176000 | consumed tokens: 4456448000 | elapsed time per iteration (s): 0.30 | learning rate: 1.409E-04 | global batch size: 256 | lm loss: 3.571654E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.213 | TFLOPs: 30.29 | 7: iteration 8510/ 21553 | consumed samples: 2178560 | consumed tokens: 4461690880 | elapsed time per iteration (s): 0.30 | learning rate: 1.408E-04 | global batch size: 256 | lm loss: 3.579580E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.187 | TFLOPs: 30.29 | 7: iteration 8520/ 21553 | consumed samples: 2181120 | consumed tokens: 4466933760 | elapsed time per iteration (s): 0.30 | learning rate: 1.407E-04 | global batch size: 256 | lm loss: 3.578230E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.952 | TFLOPs: 30.28 | 7: iteration 8530/ 21553 | consumed samples: 2183680 | consumed tokens: 4472176640 | elapsed time per iteration (s): 0.30 | learning rate: 1.406E-04 | global batch size: 256 | lm loss: 3.577899E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.374 | TFLOPs: 30.29 | 7: iteration 8540/ 21553 | consumed samples: 2186240 | consumed tokens: 4477419520 | elapsed time per iteration (s): 0.30 | learning rate: 1.405E-04 | global batch size: 256 | lm loss: 3.589219E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.863 | TFLOPs: 30.31 | 7: iteration 8550/ 21553 | consumed samples: 2188800 | consumed tokens: 4482662400 | elapsed time per iteration (s): 0.30 | learning rate: 1.403E-04 | global batch size: 256 | lm loss: 3.572705E+00 | grad norm: 0.386 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.602 | TFLOPs: 30.02 | 7: iteration 8560/ 21553 | consumed samples: 2191360 | consumed tokens: 4487905280 | elapsed time per iteration (s): 0.30 | learning rate: 1.402E-04 | global batch size: 256 | lm loss: 3.575708E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.384 | TFLOPs: 30.33 | 7: iteration 8570/ 21553 | consumed samples: 2193920 | consumed tokens: 4493148160 | elapsed time per iteration (s): 0.30 | learning rate: 1.401E-04 | global batch size: 256 | lm loss: 3.573991E+00 | grad norm: 0.387 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.293 | TFLOPs: 30.29 | 7: iteration 8580/ 21553 | consumed samples: 2196480 | consumed tokens: 4498391040 | elapsed time per iteration (s): 0.30 | learning rate: 1.400E-04 | global batch size: 256 | lm loss: 3.581570E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.587 | TFLOPs: 30.30 | 7: iteration 8590/ 21553 | consumed samples: 2199040 | consumed tokens: 4503633920 | elapsed time per iteration (s): 0.30 | learning rate: 1.398E-04 | global batch size: 256 | lm loss: 3.577329E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.054 | TFLOPs: 30.32 | 7: iteration 8600/ 21553 | consumed samples: 2201600 | consumed tokens: 4508876800 | elapsed time per iteration (s): 0.30 | learning rate: 1.397E-04 | global batch size: 256 | lm loss: 3.569780E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.107 | TFLOPs: 30.32 | 7: iteration 8610/ 21553 | consumed samples: 2204160 | consumed tokens: 4514119680 | elapsed time per iteration (s): 0.30 | learning rate: 1.396E-04 | global batch size: 256 | lm loss: 3.582884E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.091 | TFLOPs: 30.32 | 7: iteration 8620/ 21553 | consumed samples: 2206720 | consumed tokens: 4519362560 | elapsed time per iteration (s): 0.30 | learning rate: 1.395E-04 | global batch size: 256 | lm loss: 3.579821E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.020 | TFLOPs: 30.32 | 7: iteration 8630/ 21553 | consumed samples: 2209280 | consumed tokens: 4524605440 | elapsed time per iteration (s): 0.30 | learning rate: 1.393E-04 | global batch size: 256 | lm loss: 3.573611E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.790 | TFLOPs: 30.31 | 7: iteration 8640/ 21553 | consumed samples: 2211840 | consumed tokens: 4529848320 | elapsed time per iteration (s): 0.30 | learning rate: 1.392E-04 | global batch size: 256 | lm loss: 3.579448E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.001 | TFLOPs: 30.28 | 7: iteration 8650/ 21553 | consumed samples: 2214400 | consumed tokens: 4535091200 | elapsed time per iteration (s): 0.30 | learning rate: 1.391E-04 | global batch size: 256 | lm loss: 3.576729E+00 | grad norm: 0.377 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.783 | TFLOPs: 30.31 | 7: iteration 8660/ 21553 | consumed samples: 2216960 | consumed tokens: 4540334080 | elapsed time per iteration (s): 0.30 | learning rate: 1.389E-04 | global batch size: 256 | lm loss: 3.576950E+00 | grad norm: 0.397 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.704 | TFLOPs: 30.31 | 7: iteration 8670/ 21553 | consumed samples: 2219520 | consumed tokens: 4545576960 | elapsed time per iteration (s): 0.30 | learning rate: 1.388E-04 | global batch size: 256 | lm loss: 3.576279E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.495 | TFLOPs: 30.30 | 7: iteration 8680/ 21553 | consumed samples: 2222080 | consumed tokens: 4550819840 | elapsed time per iteration (s): 0.30 | learning rate: 1.387E-04 | global batch size: 256 | lm loss: 3.573793E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.098 | TFLOPs: 30.28 | 7: iteration 8690/ 21553 | consumed samples: 2224640 | consumed tokens: 4556062720 | elapsed time per iteration (s): 0.30 | learning rate: 1.386E-04 | global batch size: 256 | lm loss: 3.573336E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.068 | TFLOPs: 30.28 | 7: iteration 8700/ 21553 | consumed samples: 2227200 | consumed tokens: 4561305600 | elapsed time per iteration (s): 0.30 | learning rate: 1.384E-04 | global batch size: 256 | lm loss: 3.571795E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.947 | TFLOPs: 30.28 | 7: iteration 8710/ 21553 | consumed samples: 2229760 | consumed tokens: 4566548480 | elapsed time per iteration (s): 0.30 | learning rate: 1.383E-04 | global batch size: 256 | lm loss: 3.569126E+00 | grad norm: 0.387 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.360 | TFLOPs: 30.29 | 7: iteration 8720/ 21553 | consumed samples: 2232320 | consumed tokens: 4571791360 | elapsed time per iteration (s): 0.30 | learning rate: 1.382E-04 | global batch size: 256 | lm loss: 3.571755E+00 | grad norm: 0.392 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.360 | TFLOPs: 30.29 | 7: iteration 8730/ 21553 | consumed samples: 2234880 | consumed tokens: 4577034240 | elapsed time per iteration (s): 0.30 | learning rate: 1.381E-04 | global batch size: 256 | lm loss: 3.571950E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.558 | TFLOPs: 30.27 | 7: iteration 8740/ 21553 | consumed samples: 2237440 | consumed tokens: 4582277120 | elapsed time per iteration (s): 0.30 | learning rate: 1.379E-04 | global batch size: 256 | lm loss: 3.563336E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.696 | TFLOPs: 30.27 | 7: iteration 8750/ 21553 | consumed samples: 2240000 | consumed tokens: 4587520000 | elapsed time per iteration (s): 0.30 | learning rate: 1.378E-04 | global batch size: 256 | lm loss: 3.570945E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.734 | TFLOPs: 30.27 | 7: iteration 8760/ 21553 | consumed samples: 2242560 | consumed tokens: 4592762880 | elapsed time per iteration (s): 0.30 | learning rate: 1.377E-04 | global batch size: 256 | lm loss: 3.567293E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.856 | TFLOPs: 30.28 | 7: iteration 8770/ 21553 | consumed samples: 2245120 | consumed tokens: 4598005760 | elapsed time per iteration (s): 0.30 | learning rate: 1.376E-04 | global batch size: 256 | lm loss: 3.568256E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.461 | TFLOPs: 30.30 | 7: iteration 8780/ 21553 | consumed samples: 2247680 | consumed tokens: 4603248640 | elapsed time per iteration (s): 0.30 | learning rate: 1.374E-04 | global batch size: 256 | lm loss: 3.573045E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.093 | TFLOPs: 30.28 | 7: iteration 8790/ 21553 | consumed samples: 2250240 | consumed tokens: 4608491520 | elapsed time per iteration (s): 0.30 | learning rate: 1.373E-04 | global batch size: 256 | lm loss: 3.568635E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.116 | TFLOPs: 30.29 | 7: iteration 8800/ 21553 | consumed samples: 2252800 | consumed tokens: 4613734400 | elapsed time per iteration (s): 0.30 | learning rate: 1.372E-04 | global batch size: 256 | lm loss: 3.568801E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.788 | TFLOPs: 30.27 | 7: iteration 8810/ 21553 | consumed samples: 2255360 | consumed tokens: 4618977280 | elapsed time per iteration (s): 0.30 | learning rate: 1.371E-04 | global batch size: 256 | lm loss: 3.563627E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.989 | TFLOPs: 30.28 | 7: iteration 8820/ 21553 | consumed samples: 2257920 | consumed tokens: 4624220160 | elapsed time per iteration (s): 0.30 | learning rate: 1.369E-04 | global batch size: 256 | lm loss: 3.569887E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.486 | TFLOPs: 30.26 | 7: iteration 8830/ 21553 | consumed samples: 2260480 | consumed tokens: 4629463040 | elapsed time per iteration (s): 0.30 | learning rate: 1.368E-04 | global batch size: 256 | lm loss: 3.569557E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.204 | TFLOPs: 30.29 | 7: iteration 8840/ 21553 | consumed samples: 2263040 | consumed tokens: 4634705920 | elapsed time per iteration (s): 0.30 | learning rate: 1.367E-04 | global batch size: 256 | lm loss: 3.569268E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.683 | TFLOPs: 30.27 | 7: iteration 8850/ 21553 | consumed samples: 2265600 | consumed tokens: 4639948800 | elapsed time per iteration (s): 0.30 | learning rate: 1.366E-04 | global batch size: 256 | lm loss: 3.561835E+00 | grad norm: 0.397 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.197 | TFLOPs: 30.29 | 7: iteration 8860/ 21553 | consumed samples: 2268160 | consumed tokens: 4645191680 | elapsed time per iteration (s): 0.30 | learning rate: 1.364E-04 | global batch size: 256 | lm loss: 3.565114E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.970 | TFLOPs: 30.28 | 7: iteration 8870/ 21553 | consumed samples: 2270720 | consumed tokens: 4650434560 | elapsed time per iteration (s): 0.30 | learning rate: 1.363E-04 | global batch size: 256 | lm loss: 3.562460E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.369 | TFLOPs: 30.29 | 7: iteration 8880/ 21553 | consumed samples: 2273280 | consumed tokens: 4655677440 | elapsed time per iteration (s): 0.30 | learning rate: 1.362E-04 | global batch size: 256 | lm loss: 3.568074E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.492 | TFLOPs: 30.26 | 7: iteration 8890/ 21553 | consumed samples: 2275840 | consumed tokens: 4660920320 | elapsed time per iteration (s): 0.30 | learning rate: 1.360E-04 | global batch size: 256 | lm loss: 3.564846E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.929 | TFLOPs: 30.28 | 7: iteration 8900/ 21553 | consumed samples: 2278400 | consumed tokens: 4666163200 | elapsed time per iteration (s): 0.30 | learning rate: 1.359E-04 | global batch size: 256 | lm loss: 3.561026E+00 | grad norm: 0.376 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.216 | TFLOPs: 30.29 | 7: iteration 8910/ 21553 | consumed samples: 2280960 | consumed tokens: 4671406080 | elapsed time per iteration (s): 0.30 | learning rate: 1.358E-04 | global batch size: 256 | lm loss: 3.555318E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.630 | TFLOPs: 30.27 | 7: iteration 8920/ 21553 | consumed samples: 2283520 | consumed tokens: 4676648960 | elapsed time per iteration (s): 0.30 | learning rate: 1.357E-04 | global batch size: 256 | lm loss: 3.563240E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.249 | TFLOPs: 30.29 | 7: iteration 8930/ 21553 | consumed samples: 2286080 | consumed tokens: 4681891840 | elapsed time per iteration (s): 0.30 | learning rate: 1.355E-04 | global batch size: 256 | lm loss: 3.564355E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.208 | TFLOPs: 30.29 | 7: iteration 8940/ 21553 | consumed samples: 2288640 | consumed tokens: 4687134720 | elapsed time per iteration (s): 0.30 | learning rate: 1.354E-04 | global batch size: 256 | lm loss: 3.556527E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.006 | TFLOPs: 30.28 | 7: iteration 8950/ 21553 | consumed samples: 2291200 | consumed tokens: 4692377600 | elapsed time per iteration (s): 0.30 | learning rate: 1.353E-04 | global batch size: 256 | lm loss: 3.567774E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.164 | TFLOPs: 30.29 | 7: iteration 8960/ 21553 | consumed samples: 2293760 | consumed tokens: 4697620480 | elapsed time per iteration (s): 0.30 | learning rate: 1.352E-04 | global batch size: 256 | lm loss: 3.564743E+00 | grad norm: 0.383 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.316 | TFLOPs: 30.29 | 7: iteration 8970/ 21553 | consumed samples: 2296320 | consumed tokens: 4702863360 | elapsed time per iteration (s): 0.30 | learning rate: 1.350E-04 | global batch size: 256 | lm loss: 3.565685E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.979 | TFLOPs: 30.28 | 7: iteration 8980/ 21553 | consumed samples: 2298880 | consumed tokens: 4708106240 | elapsed time per iteration (s): 0.30 | learning rate: 1.349E-04 | global batch size: 256 | lm loss: 3.565696E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.991 | TFLOPs: 30.28 | 7: iteration 8990/ 21553 | consumed samples: 2301440 | consumed tokens: 4713349120 | elapsed time per iteration (s): 0.30 | learning rate: 1.348E-04 | global batch size: 256 | lm loss: 3.555348E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.741 | TFLOPs: 30.27 | 7: iteration 9000/ 21553 | consumed samples: 2304000 | consumed tokens: 4718592000 | elapsed time per iteration (s): 0.30 | learning rate: 1.346E-04 | global batch size: 256 | lm loss: 3.566420E+00 | grad norm: 0.482 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.782 | TFLOPs: 30.31 | 7: ----------------------------------------------------------------------------------------------- 7: validation loss at iteration 9000 | lm loss value: 3.891645E+00 | lm loss PPL: 4.899143E+01 | 7: ----------------------------------------------------------------------------------------------- 0: saving checkpoint at iteration 9000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:40:29,914] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step9000 is begin to save! 0: [2023-03-17 00:40:29,918] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:40:30,007] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:40:30,008] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:40:30,030] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:40:30,030] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:40:30,045] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:40:30,045] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:40:30,060] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:40:30,060] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:40:30,075] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:40:30,075] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:40:30,090] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:40:30,090] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:40:30,105] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:40:30,105] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:40:30,120] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:40:30,120] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:40:30,135] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:40:30,135] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:40:30,150] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:40:30,150] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:40:30,165] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:40:30,165] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:40:30,180] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:40:30,180] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:40:30,195] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:40:30,195] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:40:30,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:40:30,210] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:40:30,225] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:40:30,225] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:40:30,240] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:40:30,240] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:40:30,241] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:40:30,242] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step9000/mp_rank_00_model_states.pt 0: [2023-03-17 00:40:30,242] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:40:30,244] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:40:30,262] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:40:30,293] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:40:30,294] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:40:30,294] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:40:30,294] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: [2023-03-17 00:40:30,295] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:40:30,295] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 00:40:30,295] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 2: [2023-03-17 00:40:30,295] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:40:30,295] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:40:30,295] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 2: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:40:30,296] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 1: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 1: [2023-03-17 00:40:30,296] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 4: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:40:30,296] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 00:40:30,296] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 5: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 3: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:40:30,297] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 00:40:30,297] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 3: [2023-03-17 00:40:30,297] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:40:30,297] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:40:30,297] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 5: [2023-03-17 00:40:30,297] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:40:30,297] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:40:30,297] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: [2023-03-17 00:40:30,297] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:40:30,298] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 2: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:40:30,298] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 3: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:40:30,299] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 1: [2023-03-17 00:40:30,299] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 3: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 1: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 2: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:40:30,299] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 2: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:40:30,299] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 3: [2023-03-17 00:40:30,299] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 4: [2023-03-17 00:40:30,296] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:40:30,296] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 4: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:40:30,298] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 4: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:40:30,298] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 4: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:40:30,298] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 4: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:40:30,300] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 0: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: [2023-03-17 00:40:30,300] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 5: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:40:30,300] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 6: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:40:30,300] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 1: [2023-03-17 00:40:30,300] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 0: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 1: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 5: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:40:30,301] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:40:30,297] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:40:30,301] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 7: [2023-03-17 00:40:30,297] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 00:40:30,297] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 7: [2023-03-17 00:40:30,298] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:40:30,299] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 0: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 7: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:40:30,299] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 4: [2023-03-17 00:40:30,301] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 7: [2023-03-17 00:40:30,299] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: [2023-03-17 00:40:30,301] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 4: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 1: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:40:30,301] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:40:30,301] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 1: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 3: [2023-03-17 00:40:30,301] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:40:30,302] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 3: [2023-03-17 00:40:30,302] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 3: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:40:30,302] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 1: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:40:30,303] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 2: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:40:30,303] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:40:30,303] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 2: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:40:30,303] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 2: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:40:30,304] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:40:30,304] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 3: [2023-03-17 00:40:30,304] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:40:30,304] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 00:40:30,304] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 4: [2023-03-17 00:40:30,304] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 3: [2023-03-17 00:40:30,304] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:40:30,304] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 3: [2023-03-17 00:40:30,304] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 00:40:30,304] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 4: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:40:30,306] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 5: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:40:30,306] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 5: [2023-03-17 00:40:30,306] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 00:40:30,306] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 5: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 1: [2023-03-17 00:40:30,309] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:40:30,309] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:40:30,309] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 6: [2023-03-17 00:40:30,300] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 00:40:30,300] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 6: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:40:30,302] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:40:30,302] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 6: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 6: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:40:30,303] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:40:30,303] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 00:40:30,303] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 6: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 6: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 6: [2023-03-17 00:40:30,305] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:40:30,305] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 00:40:30,305] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 6: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:40:30,306] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 00:40:30,306] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: [2023-03-17 00:40:30,312] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 7: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:40:30,312] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 7: [2023-03-17 00:40:30,302] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 00:40:30,302] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 7: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:40:30,303] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 7: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:40:30,303] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:40:30,303] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 7: [2023-03-17 00:40:30,312] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:40:30,312] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step9000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 00:40:30,312] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step9000 is ready now! 0: successfully saved checkpoint at iteration 9000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 414.90 7: iteration 9010/ 21553 | consumed samples: 2306560 | consumed tokens: 4723834880 | elapsed time per iteration (s): 0.35 | learning rate: 1.345E-04 | global batch size: 256 | lm loss: 3.551292E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 738.674 | TFLOPs: 25.86 | 7: iteration 9020/ 21553 | consumed samples: 2309120 | consumed tokens: 4729077760 | elapsed time per iteration (s): 0.30 | learning rate: 1.344E-04 | global batch size: 256 | lm loss: 3.553325E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.375 | TFLOPs: 30.05 | 7: iteration 9030/ 21553 | consumed samples: 2311680 | consumed tokens: 4734320640 | elapsed time per iteration (s): 0.30 | learning rate: 1.343E-04 | global batch size: 256 | lm loss: 3.561029E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.839 | TFLOPs: 30.31 | 7: iteration 9040/ 21553 | consumed samples: 2314240 | consumed tokens: 4739563520 | elapsed time per iteration (s): 0.30 | learning rate: 1.341E-04 | global batch size: 256 | lm loss: 3.554565E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.495 | TFLOPs: 30.30 | 7: iteration 9050/ 21553 | consumed samples: 2316800 | consumed tokens: 4744806400 | elapsed time per iteration (s): 0.30 | learning rate: 1.340E-04 | global batch size: 256 | lm loss: 3.558278E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.561 | TFLOPs: 30.30 | 7: iteration 9060/ 21553 | consumed samples: 2319360 | consumed tokens: 4750049280 | elapsed time per iteration (s): 0.30 | learning rate: 1.339E-04 | global batch size: 256 | lm loss: 3.553352E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.761 | TFLOPs: 30.31 | 7: iteration 9070/ 21553 | consumed samples: 2321920 | consumed tokens: 4755292160 | elapsed time per iteration (s): 0.30 | learning rate: 1.338E-04 | global batch size: 256 | lm loss: 3.554447E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.329 | TFLOPs: 30.29 | 7: iteration 9080/ 21553 | consumed samples: 2324480 | consumed tokens: 4760535040 | elapsed time per iteration (s): 0.30 | learning rate: 1.336E-04 | global batch size: 256 | lm loss: 3.556529E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.587 | TFLOPs: 30.30 | 7: iteration 9090/ 21553 | consumed samples: 2327040 | consumed tokens: 4765777920 | elapsed time per iteration (s): 0.30 | learning rate: 1.335E-04 | global batch size: 256 | lm loss: 3.559997E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.699 | TFLOPs: 30.31 | 7: iteration 9100/ 21553 | consumed samples: 2329600 | consumed tokens: 4771020800 | elapsed time per iteration (s): 0.30 | learning rate: 1.334E-04 | global batch size: 256 | lm loss: 3.547362E+00 | grad norm: 0.375 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.992 | TFLOPs: 30.28 | 7: iteration 9110/ 21553 | consumed samples: 2332160 | consumed tokens: 4776263680 | elapsed time per iteration (s): 0.30 | learning rate: 1.332E-04 | global batch size: 256 | lm loss: 3.560803E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.266 | TFLOPs: 30.33 | 7: iteration 9120/ 21553 | consumed samples: 2334720 | consumed tokens: 4781506560 | elapsed time per iteration (s): 0.30 | learning rate: 1.331E-04 | global batch size: 256 | lm loss: 3.550784E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.391 | TFLOPs: 30.33 | 7: iteration 9130/ 21553 | consumed samples: 2337280 | consumed tokens: 4786749440 | elapsed time per iteration (s): 0.30 | learning rate: 1.330E-04 | global batch size: 256 | lm loss: 3.552407E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.955 | TFLOPs: 30.31 | 7: iteration 9140/ 21553 | consumed samples: 2339840 | consumed tokens: 4791992320 | elapsed time per iteration (s): 0.30 | learning rate: 1.329E-04 | global batch size: 256 | lm loss: 3.549426E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.813 | TFLOPs: 30.31 | 7: iteration 9150/ 21553 | consumed samples: 2342400 | consumed tokens: 4797235200 | elapsed time per iteration (s): 0.30 | learning rate: 1.327E-04 | global batch size: 256 | lm loss: 3.548613E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.697 | TFLOPs: 30.31 | 7: iteration 9160/ 21553 | consumed samples: 2344960 | consumed tokens: 4802478080 | elapsed time per iteration (s): 0.30 | learning rate: 1.326E-04 | global batch size: 256 | lm loss: 3.549952E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.544 | TFLOPs: 30.30 | 7: iteration 9170/ 21553 | consumed samples: 2347520 | consumed tokens: 4807720960 | elapsed time per iteration (s): 0.30 | learning rate: 1.325E-04 | global batch size: 256 | lm loss: 3.559356E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.518 | TFLOPs: 30.30 | 7: iteration 9180/ 21553 | consumed samples: 2350080 | consumed tokens: 4812963840 | elapsed time per iteration (s): 0.30 | learning rate: 1.323E-04 | global batch size: 256 | lm loss: 3.547425E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.106 | TFLOPs: 30.28 | 7: iteration 9190/ 21553 | consumed samples: 2352640 | consumed tokens: 4818206720 | elapsed time per iteration (s): 0.30 | learning rate: 1.322E-04 | global batch size: 256 | lm loss: 3.552507E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.215 | TFLOPs: 30.29 | 7: iteration 9200/ 21553 | consumed samples: 2355200 | consumed tokens: 4823449600 | elapsed time per iteration (s): 0.30 | learning rate: 1.321E-04 | global batch size: 256 | lm loss: 3.545079E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.379 | TFLOPs: 30.29 | 7: iteration 9210/ 21553 | consumed samples: 2357760 | consumed tokens: 4828692480 | elapsed time per iteration (s): 0.30 | learning rate: 1.320E-04 | global batch size: 256 | lm loss: 3.549949E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.083 | TFLOPs: 30.28 | 7: iteration 9220/ 21553 | consumed samples: 2360320 | consumed tokens: 4833935360 | elapsed time per iteration (s): 0.30 | learning rate: 1.318E-04 | global batch size: 256 | lm loss: 3.546952E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.838 | TFLOPs: 30.28 | 7: iteration 9230/ 21553 | consumed samples: 2362880 | consumed tokens: 4839178240 | elapsed time per iteration (s): 0.30 | learning rate: 1.317E-04 | global batch size: 256 | lm loss: 3.550665E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.891 | TFLOPs: 30.28 | 7: iteration 9240/ 21553 | consumed samples: 2365440 | consumed tokens: 4844421120 | elapsed time per iteration (s): 0.30 | learning rate: 1.316E-04 | global batch size: 256 | lm loss: 3.550472E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.792 | TFLOPs: 30.27 | 7: iteration 9250/ 21553 | consumed samples: 2368000 | consumed tokens: 4849664000 | elapsed time per iteration (s): 0.30 | learning rate: 1.314E-04 | global batch size: 256 | lm loss: 3.546445E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.505 | TFLOPs: 30.30 | 7: iteration 9260/ 21553 | consumed samples: 2370560 | consumed tokens: 4854906880 | elapsed time per iteration (s): 0.30 | learning rate: 1.313E-04 | global batch size: 256 | lm loss: 3.545869E+00 | grad norm: 0.457 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.141 | TFLOPs: 30.29 | 7: iteration 9270/ 21553 | consumed samples: 2373120 | consumed tokens: 4860149760 | elapsed time per iteration (s): 0.30 | learning rate: 1.312E-04 | global batch size: 256 | lm loss: 3.552878E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.610 | TFLOPs: 30.27 | 7: iteration 9280/ 21553 | consumed samples: 2375680 | consumed tokens: 4865392640 | elapsed time per iteration (s): 0.30 | learning rate: 1.311E-04 | global batch size: 256 | lm loss: 3.549569E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.845 | TFLOPs: 30.28 | 7: iteration 9290/ 21553 | consumed samples: 2378240 | consumed tokens: 4870635520 | elapsed time per iteration (s): 0.30 | learning rate: 1.309E-04 | global batch size: 256 | lm loss: 3.546372E+00 | grad norm: 0.376 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.300 | TFLOPs: 30.29 | 7: iteration 9300/ 21553 | consumed samples: 2380800 | consumed tokens: 4875878400 | elapsed time per iteration (s): 0.30 | learning rate: 1.308E-04 | global batch size: 256 | lm loss: 3.549554E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.873 | TFLOPs: 30.28 | 7: iteration 9310/ 21553 | consumed samples: 2383360 | consumed tokens: 4881121280 | elapsed time per iteration (s): 0.30 | learning rate: 1.307E-04 | global batch size: 256 | lm loss: 3.537319E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 859.844 | TFLOPs: 30.10 | 7: iteration 9320/ 21553 | consumed samples: 2385920 | consumed tokens: 4886364160 | elapsed time per iteration (s): 0.30 | learning rate: 1.305E-04 | global batch size: 256 | lm loss: 3.543608E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.686 | TFLOPs: 30.27 | 7: iteration 9330/ 21553 | consumed samples: 2388480 | consumed tokens: 4891607040 | elapsed time per iteration (s): 0.30 | learning rate: 1.304E-04 | global batch size: 256 | lm loss: 3.549291E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.092 | TFLOPs: 30.28 | 7: iteration 9340/ 21553 | consumed samples: 2391040 | consumed tokens: 4896849920 | elapsed time per iteration (s): 0.30 | learning rate: 1.303E-04 | global batch size: 256 | lm loss: 3.543127E+00 | grad norm: 0.386 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.848 | TFLOPs: 30.28 | 7: iteration 9350/ 21553 | consumed samples: 2393600 | consumed tokens: 4902092800 | elapsed time per iteration (s): 0.30 | learning rate: 1.302E-04 | global batch size: 256 | lm loss: 3.544693E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.158 | TFLOPs: 30.29 | 7: iteration 9360/ 21553 | consumed samples: 2396160 | consumed tokens: 4907335680 | elapsed time per iteration (s): 0.30 | learning rate: 1.300E-04 | global batch size: 256 | lm loss: 3.536121E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.915 | TFLOPs: 30.28 | 7: iteration 9370/ 21553 | consumed samples: 2398720 | consumed tokens: 4912578560 | elapsed time per iteration (s): 0.30 | learning rate: 1.299E-04 | global batch size: 256 | lm loss: 3.543632E+00 | grad norm: 0.385 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.605 | TFLOPs: 30.30 | 7: iteration 9380/ 21553 | consumed samples: 2401280 | consumed tokens: 4917821440 | elapsed time per iteration (s): 0.30 | learning rate: 1.298E-04 | global batch size: 256 | lm loss: 3.540070E+00 | grad norm: 0.393 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.642 | TFLOPs: 30.30 | 7: iteration 9390/ 21553 | consumed samples: 2403840 | consumed tokens: 4923064320 | elapsed time per iteration (s): 0.30 | learning rate: 1.296E-04 | global batch size: 256 | lm loss: 3.542430E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.494 | TFLOPs: 30.30 | 7: iteration 9400/ 21553 | consumed samples: 2406400 | consumed tokens: 4928307200 | elapsed time per iteration (s): 0.30 | learning rate: 1.295E-04 | global batch size: 256 | lm loss: 3.536014E+00 | grad norm: 0.387 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.622 | TFLOPs: 30.30 | 7: iteration 9410/ 21553 | consumed samples: 2408960 | consumed tokens: 4933550080 | elapsed time per iteration (s): 0.30 | learning rate: 1.294E-04 | global batch size: 256 | lm loss: 3.545234E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.120 | TFLOPs: 30.29 | 7: iteration 9420/ 21553 | consumed samples: 2411520 | consumed tokens: 4938792960 | elapsed time per iteration (s): 0.30 | learning rate: 1.293E-04 | global batch size: 256 | lm loss: 3.547980E+00 | grad norm: 0.455 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.779 | TFLOPs: 30.31 | 7: iteration 9430/ 21553 | consumed samples: 2414080 | consumed tokens: 4944035840 | elapsed time per iteration (s): 0.30 | learning rate: 1.291E-04 | global batch size: 256 | lm loss: 3.548829E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.078 | TFLOPs: 30.28 | 7: iteration 9440/ 21553 | consumed samples: 2416640 | consumed tokens: 4949278720 | elapsed time per iteration (s): 0.30 | learning rate: 1.290E-04 | global batch size: 256 | lm loss: 3.553646E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.876 | TFLOPs: 30.28 | 7: iteration 9450/ 21553 | consumed samples: 2419200 | consumed tokens: 4954521600 | elapsed time per iteration (s): 0.30 | learning rate: 1.289E-04 | global batch size: 256 | lm loss: 3.546207E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.215 | TFLOPs: 30.29 | 7: iteration 9460/ 21553 | consumed samples: 2421760 | consumed tokens: 4959764480 | elapsed time per iteration (s): 0.30 | learning rate: 1.287E-04 | global batch size: 256 | lm loss: 3.535017E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.061 | TFLOPs: 30.28 | 7: iteration 9470/ 21553 | consumed samples: 2424320 | consumed tokens: 4965007360 | elapsed time per iteration (s): 0.30 | learning rate: 1.286E-04 | global batch size: 256 | lm loss: 3.543873E+00 | grad norm: 0.373 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.089 | TFLOPs: 30.28 | 7: iteration 9480/ 21553 | consumed samples: 2426880 | consumed tokens: 4970250240 | elapsed time per iteration (s): 0.30 | learning rate: 1.285E-04 | global batch size: 256 | lm loss: 3.542866E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.862 | TFLOPs: 30.28 | 7: iteration 9490/ 21553 | consumed samples: 2429440 | consumed tokens: 4975493120 | elapsed time per iteration (s): 0.30 | learning rate: 1.283E-04 | global batch size: 256 | lm loss: 3.541625E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.867 | TFLOPs: 30.28 | 7: iteration 9500/ 21553 | consumed samples: 2432000 | consumed tokens: 4980736000 | elapsed time per iteration (s): 0.30 | learning rate: 1.282E-04 | global batch size: 256 | lm loss: 3.540718E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.406 | TFLOPs: 30.30 | 7: iteration 9510/ 21553 | consumed samples: 2434560 | consumed tokens: 4985978880 | elapsed time per iteration (s): 0.30 | learning rate: 1.281E-04 | global batch size: 256 | lm loss: 3.534132E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.325 | TFLOPs: 30.26 | 7: iteration 9520/ 21553 | consumed samples: 2437120 | consumed tokens: 4991221760 | elapsed time per iteration (s): 0.30 | learning rate: 1.280E-04 | global batch size: 256 | lm loss: 3.541623E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.055 | TFLOPs: 30.28 | 7: iteration 9530/ 21553 | consumed samples: 2439680 | consumed tokens: 4996464640 | elapsed time per iteration (s): 0.30 | learning rate: 1.278E-04 | global batch size: 256 | lm loss: 3.535077E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.007 | TFLOPs: 30.28 | 7: iteration 9540/ 21553 | consumed samples: 2442240 | consumed tokens: 5001707520 | elapsed time per iteration (s): 0.30 | learning rate: 1.277E-04 | global batch size: 256 | lm loss: 3.535698E+00 | grad norm: 0.459 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.109 | TFLOPs: 30.29 | 7: iteration 9550/ 21553 | consumed samples: 2444800 | consumed tokens: 5006950400 | elapsed time per iteration (s): 0.30 | learning rate: 1.276E-04 | global batch size: 256 | lm loss: 3.536210E+00 | grad norm: 0.393 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.943 | TFLOPs: 30.28 | 7: iteration 9560/ 21553 | consumed samples: 2447360 | consumed tokens: 5012193280 | elapsed time per iteration (s): 0.30 | learning rate: 1.274E-04 | global batch size: 256 | lm loss: 3.534396E+00 | grad norm: 0.384 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.226 | TFLOPs: 30.29 | 7: iteration 9570/ 21553 | consumed samples: 2449920 | consumed tokens: 5017436160 | elapsed time per iteration (s): 0.30 | learning rate: 1.273E-04 | global batch size: 256 | lm loss: 3.538642E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.987 | TFLOPs: 30.28 | 7: iteration 9580/ 21553 | consumed samples: 2452480 | consumed tokens: 5022679040 | elapsed time per iteration (s): 0.30 | learning rate: 1.272E-04 | global batch size: 256 | lm loss: 3.541975E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.768 | TFLOPs: 30.27 | 7: iteration 9590/ 21553 | consumed samples: 2455040 | consumed tokens: 5027921920 | elapsed time per iteration (s): 0.30 | learning rate: 1.270E-04 | global batch size: 256 | lm loss: 3.540084E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.874 | TFLOPs: 30.28 | 7: iteration 9600/ 21553 | consumed samples: 2457600 | consumed tokens: 5033164800 | elapsed time per iteration (s): 0.30 | learning rate: 1.269E-04 | global batch size: 256 | lm loss: 3.538649E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.190 | TFLOPs: 30.29 | 7: iteration 9610/ 21553 | consumed samples: 2460160 | consumed tokens: 5038407680 | elapsed time per iteration (s): 0.30 | learning rate: 1.268E-04 | global batch size: 256 | lm loss: 3.536787E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.850 | TFLOPs: 30.28 | 7: iteration 9620/ 21553 | consumed samples: 2462720 | consumed tokens: 5043650560 | elapsed time per iteration (s): 0.30 | learning rate: 1.267E-04 | global batch size: 256 | lm loss: 3.533603E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.096 | TFLOPs: 30.28 | 7: iteration 9630/ 21553 | consumed samples: 2465280 | consumed tokens: 5048893440 | elapsed time per iteration (s): 0.30 | learning rate: 1.265E-04 | global batch size: 256 | lm loss: 3.530389E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.231 | TFLOPs: 30.29 | 7: iteration 9640/ 21553 | consumed samples: 2467840 | consumed tokens: 5054136320 | elapsed time per iteration (s): 0.30 | learning rate: 1.264E-04 | global batch size: 256 | lm loss: 3.534932E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.581 | TFLOPs: 30.27 | 7: iteration 9650/ 21553 | consumed samples: 2470400 | consumed tokens: 5059379200 | elapsed time per iteration (s): 0.30 | learning rate: 1.263E-04 | global batch size: 256 | lm loss: 3.532373E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.566 | TFLOPs: 30.27 | 7: iteration 9660/ 21553 | consumed samples: 2472960 | consumed tokens: 5064622080 | elapsed time per iteration (s): 0.30 | learning rate: 1.261E-04 | global batch size: 256 | lm loss: 3.535156E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.383 | TFLOPs: 30.29 | 7: iteration 9670/ 21553 | consumed samples: 2475520 | consumed tokens: 5069864960 | elapsed time per iteration (s): 0.30 | learning rate: 1.260E-04 | global batch size: 256 | lm loss: 3.533054E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.258 | TFLOPs: 30.29 | 7: iteration 9680/ 21553 | consumed samples: 2478080 | consumed tokens: 5075107840 | elapsed time per iteration (s): 0.30 | learning rate: 1.259E-04 | global batch size: 256 | lm loss: 3.534438E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.851 | TFLOPs: 30.31 | 7: iteration 9690/ 21553 | consumed samples: 2480640 | consumed tokens: 5080350720 | elapsed time per iteration (s): 0.30 | learning rate: 1.257E-04 | global batch size: 256 | lm loss: 3.528514E+00 | grad norm: 0.459 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.268 | TFLOPs: 30.29 | 7: iteration 9700/ 21553 | consumed samples: 2483200 | consumed tokens: 5085593600 | elapsed time per iteration (s): 0.30 | learning rate: 1.256E-04 | global batch size: 256 | lm loss: 3.538863E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.649 | TFLOPs: 30.30 | 7: iteration 9710/ 21553 | consumed samples: 2485760 | consumed tokens: 5090836480 | elapsed time per iteration (s): 0.30 | learning rate: 1.255E-04 | global batch size: 256 | lm loss: 3.537721E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.975 | TFLOPs: 30.32 | 7: iteration 9720/ 21553 | consumed samples: 2488320 | consumed tokens: 5096079360 | elapsed time per iteration (s): 0.30 | learning rate: 1.254E-04 | global batch size: 256 | lm loss: 3.532854E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.963 | TFLOPs: 30.31 | 7: iteration 9730/ 21553 | consumed samples: 2490880 | consumed tokens: 5101322240 | elapsed time per iteration (s): 0.30 | learning rate: 1.252E-04 | global batch size: 256 | lm loss: 3.529060E+00 | grad norm: 0.379 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.057 | TFLOPs: 30.32 | 7: iteration 9740/ 21553 | consumed samples: 2493440 | consumed tokens: 5106565120 | elapsed time per iteration (s): 0.30 | learning rate: 1.251E-04 | global batch size: 256 | lm loss: 3.529750E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.672 | TFLOPs: 30.34 | 7: iteration 9750/ 21553 | consumed samples: 2496000 | consumed tokens: 5111808000 | elapsed time per iteration (s): 0.30 | learning rate: 1.250E-04 | global batch size: 256 | lm loss: 3.534179E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.205 | TFLOPs: 30.32 | 7: iteration 9760/ 21553 | consumed samples: 2498560 | consumed tokens: 5117050880 | elapsed time per iteration (s): 0.30 | learning rate: 1.248E-04 | global batch size: 256 | lm loss: 3.526955E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.649 | TFLOPs: 30.30 | 7: iteration 9770/ 21553 | consumed samples: 2501120 | consumed tokens: 5122293760 | elapsed time per iteration (s): 0.30 | learning rate: 1.247E-04 | global batch size: 256 | lm loss: 3.529478E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.696 | TFLOPs: 30.34 | 7: iteration 9780/ 21553 | consumed samples: 2503680 | consumed tokens: 5127536640 | elapsed time per iteration (s): 0.30 | learning rate: 1.246E-04 | global batch size: 256 | lm loss: 3.524401E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.526 | TFLOPs: 30.33 | 7: iteration 9790/ 21553 | consumed samples: 2506240 | consumed tokens: 5132779520 | elapsed time per iteration (s): 0.30 | learning rate: 1.244E-04 | global batch size: 256 | lm loss: 3.526432E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.981 | TFLOPs: 30.32 | 7: iteration 9800/ 21553 | consumed samples: 2508800 | consumed tokens: 5138022400 | elapsed time per iteration (s): 0.30 | learning rate: 1.243E-04 | global batch size: 256 | lm loss: 3.526016E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.032 | TFLOPs: 30.32 | 7: iteration 9810/ 21553 | consumed samples: 2511360 | consumed tokens: 5143265280 | elapsed time per iteration (s): 0.30 | learning rate: 1.242E-04 | global batch size: 256 | lm loss: 3.531719E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.212 | TFLOPs: 30.32 | 7: iteration 9820/ 21553 | consumed samples: 2513920 | consumed tokens: 5148508160 | elapsed time per iteration (s): 0.30 | learning rate: 1.240E-04 | global batch size: 256 | lm loss: 3.529101E+00 | grad norm: 0.458 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.898 | TFLOPs: 30.31 | 7: iteration 9830/ 21553 | consumed samples: 2516480 | consumed tokens: 5153751040 | elapsed time per iteration (s): 0.30 | learning rate: 1.239E-04 | global batch size: 256 | lm loss: 3.526810E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.983 | TFLOPs: 30.32 | 7: iteration 9840/ 21553 | consumed samples: 2519040 | consumed tokens: 5158993920 | elapsed time per iteration (s): 0.30 | learning rate: 1.238E-04 | global batch size: 256 | lm loss: 3.533135E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.506 | TFLOPs: 30.30 | 7: iteration 9850/ 21553 | consumed samples: 2521600 | consumed tokens: 5164236800 | elapsed time per iteration (s): 0.30 | learning rate: 1.237E-04 | global batch size: 256 | lm loss: 3.524934E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.413 | TFLOPs: 30.26 | 7: iteration 9860/ 21553 | consumed samples: 2524160 | consumed tokens: 5169479680 | elapsed time per iteration (s): 0.30 | learning rate: 1.235E-04 | global batch size: 256 | lm loss: 3.536076E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.260 | TFLOPs: 30.26 | 7: iteration 9870/ 21553 | consumed samples: 2526720 | consumed tokens: 5174722560 | elapsed time per iteration (s): 0.30 | learning rate: 1.234E-04 | global batch size: 256 | lm loss: 3.527396E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.462 | TFLOPs: 30.16 | 7: iteration 9880/ 21553 | consumed samples: 2529280 | consumed tokens: 5179965440 | elapsed time per iteration (s): 0.30 | learning rate: 1.233E-04 | global batch size: 256 | lm loss: 3.523289E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.008 | TFLOPs: 30.32 | 7: iteration 9890/ 21553 | consumed samples: 2531840 | consumed tokens: 5185208320 | elapsed time per iteration (s): 0.30 | learning rate: 1.231E-04 | global batch size: 256 | lm loss: 3.518380E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.510 | TFLOPs: 30.30 | 7: iteration 9900/ 21553 | consumed samples: 2534400 | consumed tokens: 5190451200 | elapsed time per iteration (s): 0.30 | learning rate: 1.230E-04 | global batch size: 256 | lm loss: 3.528043E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.402 | TFLOPs: 30.30 | 7: iteration 9910/ 21553 | consumed samples: 2536960 | consumed tokens: 5195694080 | elapsed time per iteration (s): 0.30 | learning rate: 1.229E-04 | global batch size: 256 | lm loss: 3.522578E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.180 | TFLOPs: 30.29 | 7: iteration 9920/ 21553 | consumed samples: 2539520 | consumed tokens: 5200936960 | elapsed time per iteration (s): 0.30 | learning rate: 1.227E-04 | global batch size: 256 | lm loss: 3.519282E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.439 | TFLOPs: 30.30 | 7: iteration 9930/ 21553 | consumed samples: 2542080 | consumed tokens: 5206179840 | elapsed time per iteration (s): 0.30 | learning rate: 1.226E-04 | global batch size: 256 | lm loss: 3.523029E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.509 | TFLOPs: 30.30 | 7: iteration 9940/ 21553 | consumed samples: 2544640 | consumed tokens: 5211422720 | elapsed time per iteration (s): 0.30 | learning rate: 1.225E-04 | global batch size: 256 | lm loss: 3.520884E+00 | grad norm: 0.450 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.547 | TFLOPs: 30.30 | 7: iteration 9950/ 21553 | consumed samples: 2547200 | consumed tokens: 5216665600 | elapsed time per iteration (s): 0.30 | learning rate: 1.223E-04 | global batch size: 256 | lm loss: 3.527040E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.766 | TFLOPs: 30.31 | 7: iteration 9960/ 21553 | consumed samples: 2549760 | consumed tokens: 5221908480 | elapsed time per iteration (s): 0.30 | learning rate: 1.222E-04 | global batch size: 256 | lm loss: 3.528047E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.820 | TFLOPs: 30.31 | 7: iteration 9970/ 21553 | consumed samples: 2552320 | consumed tokens: 5227151360 | elapsed time per iteration (s): 0.30 | learning rate: 1.221E-04 | global batch size: 256 | lm loss: 3.526106E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.667 | TFLOPs: 30.30 | 7: iteration 9980/ 21553 | consumed samples: 2554880 | consumed tokens: 5232394240 | elapsed time per iteration (s): 0.30 | learning rate: 1.219E-04 | global batch size: 256 | lm loss: 3.509838E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.210 | TFLOPs: 30.32 | 7: iteration 9990/ 21553 | consumed samples: 2557440 | consumed tokens: 5237637120 | elapsed time per iteration (s): 0.30 | learning rate: 1.218E-04 | global batch size: 256 | lm loss: 3.520015E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.893 | TFLOPs: 30.31 | 0: [2023-03-17 00:45:26,193] [INFO] [logging.py:68:log_dist] [Rank 0] step=10000, skipped=0, lr=[0.00012168458711439383, 0.00012168458711439383, 0.00012168458711439383], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 10000/ 21553 | consumed samples: 2560000 | consumed tokens: 5242880000 | elapsed time per iteration (s): 0.30 | learning rate: 1.217E-04 | global batch size: 256 | lm loss: 3.524731E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.704 | TFLOPs: 30.31 | 0: steps: 10000 loss: 3.5142 iter time (s): 0.294 samples/sec: 870.882 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 10000 | lm loss value: 3.860054E+00 | lm loss PPL: 4.746790E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 10000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:45:26,314] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step10000 is begin to save! 0: [2023-03-17 00:45:26,317] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:45:26,403] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:45:26,404] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:45:26,419] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:45:26,419] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:45:26,434] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:45:26,434] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:45:26,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:45:26,449] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:45:26,464] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:45:26,464] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:45:26,479] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:45:26,479] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:45:26,494] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:45:26,494] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:45:26,509] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:45:26,509] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:45:26,524] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:45:26,524] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:45:26,539] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:45:26,539] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:45:26,554] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:45:26,554] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:45:26,569] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:45:26,569] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:45:26,584] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:45:26,584] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:45:26,599] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:45:26,599] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:45:26,614] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:45:26,614] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:45:26,629] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:45:26,629] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:45:26,630] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:45:26,631] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step10000/mp_rank_00_model_states.pt 0: [2023-03-17 00:45:26,631] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:45:26,633] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:45:26,651] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:45:26,681] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:45:26,682] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:45:26,682] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: [2023-03-17 00:45:26,682] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,685] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,685] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,685] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,685] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 00:45:26,685] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:45:26,685] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:45:26,685] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 1: [2023-03-17 00:45:26,685] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 1: [2023-03-17 00:45:26,685] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 6: [2023-03-17 00:45:26,685] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:45:26,686] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:45:26,686] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 00:45:26,686] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:45:26,686] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 7: [2023-03-17 00:45:26,686] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 00:45:26,686] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 2: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,687] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:45:26,687] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:45:26,687] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 4: [2023-03-17 00:45:26,688] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:45:26,688] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 00:45:26,688] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 5: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:45:26,687] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 2: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:45:26,689] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 2: [2023-03-17 00:45:26,687] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 5: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 2: [2023-03-17 00:45:26,687] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 5: [2023-03-17 00:45:26,689] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 5: [2023-03-17 00:45:26,689] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 3: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:45:26,689] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 1: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,690] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 3: [2023-03-17 00:45:26,690] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 4: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:45:26,690] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 6: [2023-03-17 00:45:26,685] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 00:45:26,685] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 6: [2023-03-17 00:45:26,688] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:45:26,688] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:45:26,688] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:45:26,688] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:45:26,688] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 6: [2023-03-17 00:45:26,688] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 6: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:45:26,689] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 4: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:45:26,691] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 5: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:45:26,691] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:45:26,691] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 5: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 3: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:45:26,691] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 7: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:45:26,689] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 7: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:45:26,690] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 7: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:45:26,690] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 7: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:45:26,691] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:45:26,692] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 6: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:45:26,691] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 00:45:26,691] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 2: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:45:26,689] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 2: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:45:26,689] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:45:26,689] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: [2023-03-17 00:45:26,692] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 2: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:45:26,690] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:45:26,690] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 2: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:45:26,692] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 0: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 2: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 2: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:45:26,692] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 1: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,692] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 3: [2023-03-17 00:45:26,692] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 3: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:45:26,692] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 3: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 7: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 1: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 7: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 1: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 4: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 4: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 3: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 3: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 4: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:45:26,693] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 5: [2023-03-17 00:45:26,693] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:45:26,694] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 1: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:45:26,694] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 6: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:45:26,694] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 0: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:45:26,694] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 0: [2023-03-17 00:45:26,694] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 6: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 6: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 4: [2023-03-17 00:45:26,695] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:45:26,695] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 00:45:26,695] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 4: [2023-03-17 00:45:26,696] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:45:26,696] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 5: [2023-03-17 00:45:26,696] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:45:26,696] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:45:26,696] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 5: [2023-03-17 00:45:26,696] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 00:45:26,696] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 00:45:26,696] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 5: [2023-03-17 00:45:26,696] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 7: [2023-03-17 00:45:26,696] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:45:26,696] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:45:26,696] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: [2023-03-17 00:45:26,697] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:45:26,697] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 00:45:26,697] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 2: [2023-03-17 00:45:26,694] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 00:45:26,694] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: [2023-03-17 00:45:26,717] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step10000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:45:26,717] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step10000 is ready now! 0: successfully saved checkpoint at iteration 10000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 407.68 7: iteration 10010/ 21553 | consumed samples: 2562560 | consumed tokens: 5248122880 | elapsed time per iteration (s): 0.35 | learning rate: 1.216E-04 | global batch size: 256 | lm loss: 3.516735E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 736.522 | TFLOPs: 25.78 | 7: iteration 10020/ 21553 | consumed samples: 2565120 | consumed tokens: 5253365760 | elapsed time per iteration (s): 0.30 | learning rate: 1.214E-04 | global batch size: 256 | lm loss: 3.516674E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.908 | TFLOPs: 30.31 | 7: iteration 10030/ 21553 | consumed samples: 2567680 | consumed tokens: 5258608640 | elapsed time per iteration (s): 0.30 | learning rate: 1.213E-04 | global batch size: 256 | lm loss: 3.522577E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.604 | TFLOPs: 30.30 | 7: iteration 10040/ 21553 | consumed samples: 2570240 | consumed tokens: 5263851520 | elapsed time per iteration (s): 0.30 | learning rate: 1.212E-04 | global batch size: 256 | lm loss: 3.519012E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.588 | TFLOPs: 30.30 | 7: iteration 10050/ 21553 | consumed samples: 2572800 | consumed tokens: 5269094400 | elapsed time per iteration (s): 0.30 | learning rate: 1.210E-04 | global batch size: 256 | lm loss: 3.526376E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.945 | TFLOPs: 30.31 | 7: iteration 10060/ 21553 | consumed samples: 2575360 | consumed tokens: 5274337280 | elapsed time per iteration (s): 0.30 | learning rate: 1.209E-04 | global batch size: 256 | lm loss: 3.513771E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.737 | TFLOPs: 30.31 | 7: iteration 10070/ 21553 | consumed samples: 2577920 | consumed tokens: 5279580160 | elapsed time per iteration (s): 0.30 | learning rate: 1.208E-04 | global batch size: 256 | lm loss: 3.517965E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.661 | TFLOPs: 30.30 | 7: iteration 10080/ 21553 | consumed samples: 2580480 | consumed tokens: 5284823040 | elapsed time per iteration (s): 0.30 | learning rate: 1.206E-04 | global batch size: 256 | lm loss: 3.515398E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.685 | TFLOPs: 30.31 | 7: iteration 10090/ 21553 | consumed samples: 2583040 | consumed tokens: 5290065920 | elapsed time per iteration (s): 0.30 | learning rate: 1.205E-04 | global batch size: 256 | lm loss: 3.524790E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.627 | TFLOPs: 30.30 | 7: iteration 10100/ 21553 | consumed samples: 2585600 | consumed tokens: 5295308800 | elapsed time per iteration (s): 0.30 | learning rate: 1.204E-04 | global batch size: 256 | lm loss: 3.516793E+00 | grad norm: 0.391 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.858 | TFLOPs: 30.31 | 7: iteration 10110/ 21553 | consumed samples: 2588160 | consumed tokens: 5300551680 | elapsed time per iteration (s): 0.30 | learning rate: 1.202E-04 | global batch size: 256 | lm loss: 3.516680E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.046 | TFLOPs: 30.14 | 7: iteration 10120/ 21553 | consumed samples: 2590720 | consumed tokens: 5305794560 | elapsed time per iteration (s): 0.30 | learning rate: 1.201E-04 | global batch size: 256 | lm loss: 3.513533E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.498 | TFLOPs: 30.16 | 7: iteration 10130/ 21553 | consumed samples: 2593280 | consumed tokens: 5311037440 | elapsed time per iteration (s): 0.30 | learning rate: 1.200E-04 | global batch size: 256 | lm loss: 3.514487E+00 | grad norm: 0.444 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.530 | TFLOPs: 30.30 | 7: iteration 10140/ 21553 | consumed samples: 2595840 | consumed tokens: 5316280320 | elapsed time per iteration (s): 0.30 | learning rate: 1.198E-04 | global batch size: 256 | lm loss: 3.514075E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.333 | TFLOPs: 30.29 | 7: iteration 10150/ 21553 | consumed samples: 2598400 | consumed tokens: 5321523200 | elapsed time per iteration (s): 0.30 | learning rate: 1.197E-04 | global batch size: 256 | lm loss: 3.522710E+00 | grad norm: 0.379 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.432 | TFLOPs: 30.30 | 7: iteration 10160/ 21553 | consumed samples: 2600960 | consumed tokens: 5326766080 | elapsed time per iteration (s): 0.30 | learning rate: 1.196E-04 | global batch size: 256 | lm loss: 3.506496E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.342 | TFLOPs: 30.29 | 7: iteration 10170/ 21553 | consumed samples: 2603520 | consumed tokens: 5332008960 | elapsed time per iteration (s): 0.30 | learning rate: 1.194E-04 | global batch size: 256 | lm loss: 3.518482E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.560 | TFLOPs: 30.30 | 7: iteration 10180/ 21553 | consumed samples: 2606080 | consumed tokens: 5337251840 | elapsed time per iteration (s): 0.30 | learning rate: 1.193E-04 | global batch size: 256 | lm loss: 3.511689E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.414 | TFLOPs: 30.30 | 7: iteration 10190/ 21553 | consumed samples: 2608640 | consumed tokens: 5342494720 | elapsed time per iteration (s): 0.30 | learning rate: 1.192E-04 | global batch size: 256 | lm loss: 3.511574E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.950 | TFLOPs: 30.28 | 7: iteration 10200/ 21553 | consumed samples: 2611200 | consumed tokens: 5347737600 | elapsed time per iteration (s): 0.30 | learning rate: 1.191E-04 | global batch size: 256 | lm loss: 3.513280E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.468 | TFLOPs: 30.30 | 7: iteration 10210/ 21553 | consumed samples: 2613760 | consumed tokens: 5352980480 | elapsed time per iteration (s): 0.30 | learning rate: 1.189E-04 | global batch size: 256 | lm loss: 3.511491E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.280 | TFLOPs: 30.29 | 7: iteration 10220/ 21553 | consumed samples: 2616320 | consumed tokens: 5358223360 | elapsed time per iteration (s): 0.30 | learning rate: 1.188E-04 | global batch size: 256 | lm loss: 3.514127E+00 | grad norm: 0.384 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.453 | TFLOPs: 30.30 | 7: iteration 10230/ 21553 | consumed samples: 2618880 | consumed tokens: 5363466240 | elapsed time per iteration (s): 0.30 | learning rate: 1.187E-04 | global batch size: 256 | lm loss: 3.514742E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.670 | TFLOPs: 30.30 | 7: iteration 10240/ 21553 | consumed samples: 2621440 | consumed tokens: 5368709120 | elapsed time per iteration (s): 0.30 | learning rate: 1.185E-04 | global batch size: 256 | lm loss: 3.511101E+00 | grad norm: 0.380 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.445 | TFLOPs: 30.30 | 7: iteration 10250/ 21553 | consumed samples: 2624000 | consumed tokens: 5373952000 | elapsed time per iteration (s): 0.30 | learning rate: 1.184E-04 | global batch size: 256 | lm loss: 3.505062E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.512 | TFLOPs: 30.30 | 7: iteration 10260/ 21553 | consumed samples: 2626560 | consumed tokens: 5379194880 | elapsed time per iteration (s): 0.30 | learning rate: 1.183E-04 | global batch size: 256 | lm loss: 3.516508E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.555 | TFLOPs: 30.30 | 7: iteration 10270/ 21553 | consumed samples: 2629120 | consumed tokens: 5384437760 | elapsed time per iteration (s): 0.30 | learning rate: 1.181E-04 | global batch size: 256 | lm loss: 3.513004E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.764 | TFLOPs: 30.31 | 7: iteration 10280/ 21553 | consumed samples: 2631680 | consumed tokens: 5389680640 | elapsed time per iteration (s): 0.30 | learning rate: 1.180E-04 | global batch size: 256 | lm loss: 3.509012E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.886 | TFLOPs: 30.31 | 7: iteration 10290/ 21553 | consumed samples: 2634240 | consumed tokens: 5394923520 | elapsed time per iteration (s): 0.30 | learning rate: 1.179E-04 | global batch size: 256 | lm loss: 3.504193E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.514 | TFLOPs: 30.30 | 7: iteration 10300/ 21553 | consumed samples: 2636800 | consumed tokens: 5400166400 | elapsed time per iteration (s): 0.30 | learning rate: 1.177E-04 | global batch size: 256 | lm loss: 3.506854E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.775 | TFLOPs: 30.31 | 7: iteration 10310/ 21553 | consumed samples: 2639360 | consumed tokens: 5405409280 | elapsed time per iteration (s): 0.30 | learning rate: 1.176E-04 | global batch size: 256 | lm loss: 3.507057E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.726 | TFLOPs: 30.31 | 7: iteration 10320/ 21553 | consumed samples: 2641920 | consumed tokens: 5410652160 | elapsed time per iteration (s): 0.30 | learning rate: 1.175E-04 | global batch size: 256 | lm loss: 3.512515E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.947 | TFLOPs: 30.31 | 7: iteration 10330/ 21553 | consumed samples: 2644480 | consumed tokens: 5415895040 | elapsed time per iteration (s): 0.30 | learning rate: 1.173E-04 | global batch size: 256 | lm loss: 3.503014E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.949 | TFLOPs: 30.31 | 7: iteration 10340/ 21553 | consumed samples: 2647040 | consumed tokens: 5421137920 | elapsed time per iteration (s): 0.30 | learning rate: 1.172E-04 | global batch size: 256 | lm loss: 3.509780E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.793 | TFLOPs: 30.31 | 7: iteration 10350/ 21553 | consumed samples: 2649600 | consumed tokens: 5426380800 | elapsed time per iteration (s): 0.30 | learning rate: 1.171E-04 | global batch size: 256 | lm loss: 3.504023E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.854 | TFLOPs: 30.28 | 7: iteration 10360/ 21553 | consumed samples: 2652160 | consumed tokens: 5431623680 | elapsed time per iteration (s): 0.30 | learning rate: 1.169E-04 | global batch size: 256 | lm loss: 3.516308E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.190 | TFLOPs: 30.29 | 7: iteration 10370/ 21553 | consumed samples: 2654720 | consumed tokens: 5436866560 | elapsed time per iteration (s): 0.30 | learning rate: 1.168E-04 | global batch size: 256 | lm loss: 3.506296E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.134 | TFLOPs: 30.29 | 7: iteration 10380/ 21553 | consumed samples: 2657280 | consumed tokens: 5442109440 | elapsed time per iteration (s): 0.30 | learning rate: 1.167E-04 | global batch size: 256 | lm loss: 3.510888E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.069 | TFLOPs: 30.28 | 7: iteration 10390/ 21553 | consumed samples: 2659840 | consumed tokens: 5447352320 | elapsed time per iteration (s): 0.30 | learning rate: 1.165E-04 | global batch size: 256 | lm loss: 3.505008E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.126 | TFLOPs: 30.29 | 7: iteration 10400/ 21553 | consumed samples: 2662400 | consumed tokens: 5452595200 | elapsed time per iteration (s): 0.30 | learning rate: 1.164E-04 | global batch size: 256 | lm loss: 3.506435E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.839 | TFLOPs: 30.28 | 7: iteration 10410/ 21553 | consumed samples: 2664960 | consumed tokens: 5457838080 | elapsed time per iteration (s): 0.30 | learning rate: 1.163E-04 | global batch size: 256 | lm loss: 3.501737E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.333 | TFLOPs: 30.26 | 7: iteration 10420/ 21553 | consumed samples: 2667520 | consumed tokens: 5463080960 | elapsed time per iteration (s): 0.30 | learning rate: 1.161E-04 | global batch size: 256 | lm loss: 3.510984E+00 | grad norm: 0.390 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.123 | TFLOPs: 30.29 | 7: iteration 10430/ 21553 | consumed samples: 2670080 | consumed tokens: 5468323840 | elapsed time per iteration (s): 0.30 | learning rate: 1.160E-04 | global batch size: 256 | lm loss: 3.498232E+00 | grad norm: 0.478 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.262 | TFLOPs: 30.29 | 7: iteration 10440/ 21553 | consumed samples: 2672640 | consumed tokens: 5473566720 | elapsed time per iteration (s): 0.30 | learning rate: 1.159E-04 | global batch size: 256 | lm loss: 3.502727E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.593 | TFLOPs: 30.27 | 7: iteration 10450/ 21553 | consumed samples: 2675200 | consumed tokens: 5478809600 | elapsed time per iteration (s): 0.30 | learning rate: 1.158E-04 | global batch size: 256 | lm loss: 3.504965E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.939 | TFLOPs: 30.24 | 7: iteration 10460/ 21553 | consumed samples: 2677760 | consumed tokens: 5484052480 | elapsed time per iteration (s): 0.30 | learning rate: 1.156E-04 | global batch size: 256 | lm loss: 3.499768E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.044 | TFLOPs: 30.25 | 7: iteration 10470/ 21553 | consumed samples: 2680320 | consumed tokens: 5489295360 | elapsed time per iteration (s): 0.30 | learning rate: 1.155E-04 | global batch size: 256 | lm loss: 3.503313E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.791 | TFLOPs: 30.24 | 7: iteration 10480/ 21553 | consumed samples: 2682880 | consumed tokens: 5494538240 | elapsed time per iteration (s): 0.30 | learning rate: 1.154E-04 | global batch size: 256 | lm loss: 3.508331E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.098 | TFLOPs: 30.28 | 7: iteration 10490/ 21553 | consumed samples: 2685440 | consumed tokens: 5499781120 | elapsed time per iteration (s): 0.30 | learning rate: 1.152E-04 | global batch size: 256 | lm loss: 3.500496E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.594 | TFLOPs: 30.27 | 7: iteration 10500/ 21553 | consumed samples: 2688000 | consumed tokens: 5505024000 | elapsed time per iteration (s): 0.30 | learning rate: 1.151E-04 | global batch size: 256 | lm loss: 3.492548E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.978 | TFLOPs: 30.25 | 7: iteration 10510/ 21553 | consumed samples: 2690560 | consumed tokens: 5510266880 | elapsed time per iteration (s): 0.30 | learning rate: 1.150E-04 | global batch size: 256 | lm loss: 3.496623E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.980 | TFLOPs: 30.28 | 7: iteration 10520/ 21553 | consumed samples: 2693120 | consumed tokens: 5515509760 | elapsed time per iteration (s): 0.30 | learning rate: 1.148E-04 | global batch size: 256 | lm loss: 3.504151E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.795 | TFLOPs: 30.24 | 7: iteration 10530/ 21553 | consumed samples: 2695680 | consumed tokens: 5520752640 | elapsed time per iteration (s): 0.30 | learning rate: 1.147E-04 | global batch size: 256 | lm loss: 3.505011E+00 | grad norm: 0.449 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.300 | TFLOPs: 30.26 | 7: iteration 10540/ 21553 | consumed samples: 2698240 | consumed tokens: 5525995520 | elapsed time per iteration (s): 0.30 | learning rate: 1.146E-04 | global batch size: 256 | lm loss: 3.499404E+00 | grad norm: 0.457 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.874 | TFLOPs: 30.28 | 7: iteration 10550/ 21553 | consumed samples: 2700800 | consumed tokens: 5531238400 | elapsed time per iteration (s): 0.30 | learning rate: 1.144E-04 | global batch size: 256 | lm loss: 3.504227E+00 | grad norm: 0.461 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.674 | TFLOPs: 30.27 | 7: iteration 10560/ 21553 | consumed samples: 2703360 | consumed tokens: 5536481280 | elapsed time per iteration (s): 0.30 | learning rate: 1.143E-04 | global batch size: 256 | lm loss: 3.499112E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.536 | TFLOPs: 30.27 | 7: iteration 10570/ 21553 | consumed samples: 2705920 | consumed tokens: 5541724160 | elapsed time per iteration (s): 0.30 | learning rate: 1.142E-04 | global batch size: 256 | lm loss: 3.501421E+00 | grad norm: 0.387 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.579 | TFLOPs: 30.27 | 7: iteration 10580/ 21553 | consumed samples: 2708480 | consumed tokens: 5546967040 | elapsed time per iteration (s): 0.30 | learning rate: 1.140E-04 | global batch size: 256 | lm loss: 3.502771E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.520 | TFLOPs: 30.23 | 7: iteration 10590/ 21553 | consumed samples: 2711040 | consumed tokens: 5552209920 | elapsed time per iteration (s): 0.30 | learning rate: 1.139E-04 | global batch size: 256 | lm loss: 3.496125E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.752 | TFLOPs: 30.06 | 7: iteration 10600/ 21553 | consumed samples: 2713600 | consumed tokens: 5557452800 | elapsed time per iteration (s): 0.30 | learning rate: 1.138E-04 | global batch size: 256 | lm loss: 3.497691E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.568 | TFLOPs: 30.27 | 7: iteration 10610/ 21553 | consumed samples: 2716160 | consumed tokens: 5562695680 | elapsed time per iteration (s): 0.30 | learning rate: 1.136E-04 | global batch size: 256 | lm loss: 3.499887E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.640 | TFLOPs: 30.27 | 7: iteration 10620/ 21553 | consumed samples: 2718720 | consumed tokens: 5567938560 | elapsed time per iteration (s): 0.30 | learning rate: 1.135E-04 | global batch size: 256 | lm loss: 3.503770E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.031 | TFLOPs: 30.21 | 7: iteration 10630/ 21553 | consumed samples: 2721280 | consumed tokens: 5573181440 | elapsed time per iteration (s): 0.30 | learning rate: 1.134E-04 | global batch size: 256 | lm loss: 3.491137E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.406 | TFLOPs: 30.23 | 7: iteration 10640/ 21553 | consumed samples: 2723840 | consumed tokens: 5578424320 | elapsed time per iteration (s): 0.30 | learning rate: 1.132E-04 | global batch size: 256 | lm loss: 3.498835E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.675 | TFLOPs: 30.23 | 7: iteration 10650/ 21553 | consumed samples: 2726400 | consumed tokens: 5583667200 | elapsed time per iteration (s): 0.30 | learning rate: 1.131E-04 | global batch size: 256 | lm loss: 3.503651E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.323 | TFLOPs: 30.19 | 7: iteration 10660/ 21553 | consumed samples: 2728960 | consumed tokens: 5588910080 | elapsed time per iteration (s): 0.30 | learning rate: 1.130E-04 | global batch size: 256 | lm loss: 3.498415E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.940 | TFLOPs: 30.24 | 7: iteration 10670/ 21553 | consumed samples: 2731520 | consumed tokens: 5594152960 | elapsed time per iteration (s): 0.30 | learning rate: 1.128E-04 | global batch size: 256 | lm loss: 3.498066E+00 | grad norm: 0.397 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.930 | TFLOPs: 30.21 | 7: iteration 10680/ 21553 | consumed samples: 2734080 | consumed tokens: 5599395840 | elapsed time per iteration (s): 0.30 | learning rate: 1.127E-04 | global batch size: 256 | lm loss: 3.499907E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.212 | TFLOPs: 30.18 | 7: iteration 10690/ 21553 | consumed samples: 2736640 | consumed tokens: 5604638720 | elapsed time per iteration (s): 0.30 | learning rate: 1.126E-04 | global batch size: 256 | lm loss: 3.500683E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.508 | TFLOPs: 30.23 | 7: iteration 10700/ 21553 | consumed samples: 2739200 | consumed tokens: 5609881600 | elapsed time per iteration (s): 0.30 | learning rate: 1.124E-04 | global batch size: 256 | lm loss: 3.492822E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.030 | TFLOPs: 30.25 | 7: iteration 10710/ 21553 | consumed samples: 2741760 | consumed tokens: 5615124480 | elapsed time per iteration (s): 0.30 | learning rate: 1.123E-04 | global batch size: 256 | lm loss: 3.490812E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.456 | TFLOPs: 30.26 | 7: iteration 10720/ 21553 | consumed samples: 2744320 | consumed tokens: 5620367360 | elapsed time per iteration (s): 0.30 | learning rate: 1.122E-04 | global batch size: 256 | lm loss: 3.494897E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.126 | TFLOPs: 30.25 | 7: iteration 10730/ 21553 | consumed samples: 2746880 | consumed tokens: 5625610240 | elapsed time per iteration (s): 0.30 | learning rate: 1.120E-04 | global batch size: 256 | lm loss: 3.499796E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.332 | TFLOPs: 30.29 | 7: iteration 10740/ 21553 | consumed samples: 2749440 | consumed tokens: 5630853120 | elapsed time per iteration (s): 0.30 | learning rate: 1.119E-04 | global batch size: 256 | lm loss: 3.489297E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.231 | TFLOPs: 30.25 | 7: iteration 10750/ 21553 | consumed samples: 2752000 | consumed tokens: 5636096000 | elapsed time per iteration (s): 0.30 | learning rate: 1.118E-04 | global batch size: 256 | lm loss: 3.496488E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.883 | TFLOPs: 30.17 | 7: iteration 10760/ 21553 | consumed samples: 2754560 | consumed tokens: 5641338880 | elapsed time per iteration (s): 0.30 | learning rate: 1.116E-04 | global batch size: 256 | lm loss: 3.490328E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.456 | TFLOPs: 30.19 | 7: iteration 10770/ 21553 | consumed samples: 2757120 | consumed tokens: 5646581760 | elapsed time per iteration (s): 0.30 | learning rate: 1.115E-04 | global batch size: 256 | lm loss: 3.492595E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.628 | TFLOPs: 30.27 | 7: iteration 10780/ 21553 | consumed samples: 2759680 | consumed tokens: 5651824640 | elapsed time per iteration (s): 0.30 | learning rate: 1.114E-04 | global batch size: 256 | lm loss: 3.495354E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.640 | TFLOPs: 30.23 | 7: iteration 10790/ 21553 | consumed samples: 2762240 | consumed tokens: 5657067520 | elapsed time per iteration (s): 0.30 | learning rate: 1.112E-04 | global batch size: 256 | lm loss: 3.489422E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.054 | TFLOPs: 30.18 | 7: iteration 10800/ 21553 | consumed samples: 2764800 | consumed tokens: 5662310400 | elapsed time per iteration (s): 0.30 | learning rate: 1.111E-04 | global batch size: 256 | lm loss: 3.496968E+00 | grad norm: 0.388 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.572 | TFLOPs: 30.27 | 7: iteration 10810/ 21553 | consumed samples: 2767360 | consumed tokens: 5667553280 | elapsed time per iteration (s): 0.30 | learning rate: 1.110E-04 | global batch size: 256 | lm loss: 3.495902E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.708 | TFLOPs: 30.27 | 7: iteration 10820/ 21553 | consumed samples: 2769920 | consumed tokens: 5672796160 | elapsed time per iteration (s): 0.30 | learning rate: 1.109E-04 | global batch size: 256 | lm loss: 3.491862E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.028 | TFLOPs: 30.25 | 7: iteration 10830/ 21553 | consumed samples: 2772480 | consumed tokens: 5678039040 | elapsed time per iteration (s): 0.30 | learning rate: 1.107E-04 | global batch size: 256 | lm loss: 3.493425E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.883 | TFLOPs: 30.28 | 7: iteration 10840/ 21553 | consumed samples: 2775040 | consumed tokens: 5683281920 | elapsed time per iteration (s): 0.30 | learning rate: 1.106E-04 | global batch size: 256 | lm loss: 3.492705E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.844 | TFLOPs: 30.28 | 7: iteration 10850/ 21553 | consumed samples: 2777600 | consumed tokens: 5688524800 | elapsed time per iteration (s): 0.30 | learning rate: 1.105E-04 | global batch size: 256 | lm loss: 3.499017E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.366 | TFLOPs: 30.26 | 7: iteration 10860/ 21553 | consumed samples: 2780160 | consumed tokens: 5693767680 | elapsed time per iteration (s): 0.30 | learning rate: 1.103E-04 | global batch size: 256 | lm loss: 3.489922E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.423 | TFLOPs: 30.26 | 7: iteration 10870/ 21553 | consumed samples: 2782720 | consumed tokens: 5699010560 | elapsed time per iteration (s): 0.30 | learning rate: 1.102E-04 | global batch size: 256 | lm loss: 3.490685E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.489 | TFLOPs: 30.30 | 7: iteration 10880/ 21553 | consumed samples: 2785280 | consumed tokens: 5704253440 | elapsed time per iteration (s): 0.30 | learning rate: 1.101E-04 | global batch size: 256 | lm loss: 3.493491E+00 | grad norm: 0.481 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.800 | TFLOPs: 30.31 | 7: iteration 10890/ 21553 | consumed samples: 2787840 | consumed tokens: 5709496320 | elapsed time per iteration (s): 0.30 | learning rate: 1.099E-04 | global batch size: 256 | lm loss: 3.485141E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.086 | TFLOPs: 30.32 | 7: iteration 10900/ 21553 | consumed samples: 2790400 | consumed tokens: 5714739200 | elapsed time per iteration (s): 0.30 | learning rate: 1.098E-04 | global batch size: 256 | lm loss: 3.493800E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.600 | TFLOPs: 30.30 | 7: iteration 10910/ 21553 | consumed samples: 2792960 | consumed tokens: 5719982080 | elapsed time per iteration (s): 0.30 | learning rate: 1.097E-04 | global batch size: 256 | lm loss: 3.485292E+00 | grad norm: 0.393 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.258 | TFLOPs: 30.26 | 7: iteration 10920/ 21553 | consumed samples: 2795520 | consumed tokens: 5725224960 | elapsed time per iteration (s): 0.30 | learning rate: 1.095E-04 | global batch size: 256 | lm loss: 3.498424E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.308 | TFLOPs: 30.26 | 7: iteration 10930/ 21553 | consumed samples: 2798080 | consumed tokens: 5730467840 | elapsed time per iteration (s): 0.30 | learning rate: 1.094E-04 | global batch size: 256 | lm loss: 3.489441E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.581 | TFLOPs: 30.27 | 7: iteration 10940/ 21553 | consumed samples: 2800640 | consumed tokens: 5735710720 | elapsed time per iteration (s): 0.32 | learning rate: 1.093E-04 | global batch size: 256 | lm loss: 3.486493E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 809.996 | TFLOPs: 28.36 | 7: iteration 10950/ 21553 | consumed samples: 2803200 | consumed tokens: 5740953600 | elapsed time per iteration (s): 0.30 | learning rate: 1.091E-04 | global batch size: 256 | lm loss: 3.496749E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.852 | TFLOPs: 30.24 | 7: iteration 10960/ 21553 | consumed samples: 2805760 | consumed tokens: 5746196480 | elapsed time per iteration (s): 0.30 | learning rate: 1.090E-04 | global batch size: 256 | lm loss: 3.488360E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.961 | TFLOPs: 30.31 | 7: iteration 10970/ 21553 | consumed samples: 2808320 | consumed tokens: 5751439360 | elapsed time per iteration (s): 0.30 | learning rate: 1.089E-04 | global batch size: 256 | lm loss: 3.492926E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.332 | TFLOPs: 30.29 | 7: iteration 10980/ 21553 | consumed samples: 2810880 | consumed tokens: 5756682240 | elapsed time per iteration (s): 0.30 | learning rate: 1.087E-04 | global batch size: 256 | lm loss: 3.491513E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.708 | TFLOPs: 30.31 | 7: iteration 10990/ 21553 | consumed samples: 2813440 | consumed tokens: 5761925120 | elapsed time per iteration (s): 0.30 | learning rate: 1.086E-04 | global batch size: 256 | lm loss: 3.489741E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.611 | TFLOPs: 30.30 | 7: iteration 11000/ 21553 | consumed samples: 2816000 | consumed tokens: 5767168000 | elapsed time per iteration (s): 0.30 | learning rate: 1.085E-04 | global batch size: 256 | lm loss: 3.478345E+00 | grad norm: 0.397 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.854 | TFLOPs: 30.31 | 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 11000 | lm loss value: 3.967298E+00 | lm loss PPL: 5.284154E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 11000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:50:23,083] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step11000 is begin to save! 0: [2023-03-17 00:50:23,087] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:50:23,172] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:50:23,172] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:50:23,188] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:50:23,189] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:50:23,203] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:50:23,204] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:50:23,218] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:50:23,218] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:50:23,233] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:50:23,233] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:50:23,248] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:50:23,248] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:50:23,263] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:50:23,263] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:50:23,278] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:50:23,278] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:50:23,293] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:50:23,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:50:23,308] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:50:23,308] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:50:23,323] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:50:23,323] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:50:23,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:50:23,338] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:50:23,353] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:50:23,353] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:50:23,368] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:50:23,368] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:50:23,383] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:50:23,383] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:50:23,398] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:50:23,398] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:50:23,399] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:50:23,400] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step11000/mp_rank_00_model_states.pt 0: [2023-03-17 00:50:23,400] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:50:23,402] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:50:23,418] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:50:23,449] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:50:23,449] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:50:23,449] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 3: [2023-03-17 00:50:23,451] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:50:23,451] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 00:50:23,451] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 2: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:50:23,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 6: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:50:23,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 1: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:50:23,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 7: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:50:23,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 6: [2023-03-17 00:50:23,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:50:23,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:50:23,453] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 0: [2023-03-17 00:50:23,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:50:23,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 00:50:23,453] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 5: [2023-03-17 00:50:23,453] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:50:23,453] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:50:23,453] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 5: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:50:23,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 7: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:50:23,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 1: [2023-03-17 00:50:23,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 7: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:50:23,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 3: [2023-03-17 00:50:23,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:50:23,455] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 00:50:23,455] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 6: [2023-03-17 00:50:23,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:50:23,455] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 6: [2023-03-17 00:50:23,455] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 2: [2023-03-17 00:50:23,452] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,452] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 2: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:50:23,454] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,454] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 2: [2023-03-17 00:50:23,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:50:23,455] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:50:23,455] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 3: [2023-03-17 00:50:23,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 3: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 0: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:50:23,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 6: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 5: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:50:23,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 0: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 0: [2023-03-17 00:50:23,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 3: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:50:23,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 6: [2023-03-17 00:50:23,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 7: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:50:23,456] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 7: [2023-03-17 00:50:23,456] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:50:23,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 4: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 4: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 7: [2023-03-17 00:50:23,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 4: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 7: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 4: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 5: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 1: [2023-03-17 00:50:23,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:50:23,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 1: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 5: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 1: [2023-03-17 00:50:23,458] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:50:23,458] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:50:23,458] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 6: [2023-03-17 00:50:23,458] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:50:23,458] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:50:23,458] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 1: [2023-03-17 00:50:23,458] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:50:23,458] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:50:23,458] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 4: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,459] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 2: [2023-03-17 00:50:23,457] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,457] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 2: [2023-03-17 00:50:23,458] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:50:23,458] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,458] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 3: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:50:23,459] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 0: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:50:23,459] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 0: [2023-03-17 00:50:23,459] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 3: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 3: [2023-03-17 00:50:23,459] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:50:23,459] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 4: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,460] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 4: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,460] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:50:23,460] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 4: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 3: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:50:23,460] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 1: [2023-03-17 00:50:23,461] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:50:23,461] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 0: [2023-03-17 00:50:23,461] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:50:23,461] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 0: [2023-03-17 00:50:23,461] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 00:50:23,461] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 1: [2023-03-17 00:50:23,461] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:50:23,461] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:50:23,461] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:50:23,461] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 1: [2023-03-17 00:50:23,461] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:50:23,461] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 4: [2023-03-17 00:50:23,462] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,463] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 7: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:50:23,463] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 7: [2023-03-17 00:50:23,460] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 7: [2023-03-17 00:50:23,463] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:50:23,463] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 00:50:23,463] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 2: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:50:23,460] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,460] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 2: [2023-03-17 00:50:23,463] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:50:23,463] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:50:23,463] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 6: [2023-03-17 00:50:23,463] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:50:23,463] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 00:50:23,463] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 6: [2023-03-17 00:50:23,464] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:50:23,464] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:50:23,464] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:50:23,464] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 6: [2023-03-17 00:50:23,464] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 00:50:23,464] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 5: [2023-03-17 00:50:23,465] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:50:23,465] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:50:23,465] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:50:23,465] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:50:23,465] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 00:50:23,465] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:50:23,465] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:50:23,465] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 00:50:23,465] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 5: [2023-03-17 00:50:23,465] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 5: [2023-03-17 00:50:23,465] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 5: [2023-03-17 00:50:23,465] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 0: [2023-03-17 00:50:23,478] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step11000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:50:23,478] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step11000 is ready now! 0: successfully saved checkpoint at iteration 11000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 401.68 7: iteration 11010/ 21553 | consumed samples: 2818560 | consumed tokens: 5772410880 | elapsed time per iteration (s): 0.35 | learning rate: 1.083E-04 | global batch size: 256 | lm loss: 3.491036E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 741.601 | TFLOPs: 25.96 | 7: iteration 11020/ 21553 | consumed samples: 2821120 | consumed tokens: 5777653760 | elapsed time per iteration (s): 0.30 | learning rate: 1.082E-04 | global batch size: 256 | lm loss: 3.488125E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.936 | TFLOPs: 30.31 | 7: iteration 11030/ 21553 | consumed samples: 2823680 | consumed tokens: 5782896640 | elapsed time per iteration (s): 0.30 | learning rate: 1.081E-04 | global batch size: 256 | lm loss: 3.485017E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.995 | TFLOPs: 30.32 | 7: iteration 11040/ 21553 | consumed samples: 2826240 | consumed tokens: 5788139520 | elapsed time per iteration (s): 0.30 | learning rate: 1.079E-04 | global batch size: 256 | lm loss: 3.488586E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.958 | TFLOPs: 30.31 | 7: iteration 11050/ 21553 | consumed samples: 2828800 | consumed tokens: 5793382400 | elapsed time per iteration (s): 0.30 | learning rate: 1.078E-04 | global batch size: 256 | lm loss: 3.485895E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.724 | TFLOPs: 30.31 | 7: iteration 11060/ 21553 | consumed samples: 2831360 | consumed tokens: 5798625280 | elapsed time per iteration (s): 0.30 | learning rate: 1.077E-04 | global batch size: 256 | lm loss: 3.483577E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.791 | TFLOPs: 30.31 | 7: iteration 11070/ 21553 | consumed samples: 2833920 | consumed tokens: 5803868160 | elapsed time per iteration (s): 0.30 | learning rate: 1.075E-04 | global batch size: 256 | lm loss: 3.488807E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.632 | TFLOPs: 30.30 | 7: iteration 11080/ 21553 | consumed samples: 2836480 | consumed tokens: 5809111040 | elapsed time per iteration (s): 0.30 | learning rate: 1.074E-04 | global batch size: 256 | lm loss: 3.484365E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.451 | TFLOPs: 30.30 | 7: iteration 11090/ 21553 | consumed samples: 2839040 | consumed tokens: 5814353920 | elapsed time per iteration (s): 0.30 | learning rate: 1.073E-04 | global batch size: 256 | lm loss: 3.482489E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.977 | TFLOPs: 30.28 | 7: iteration 11100/ 21553 | consumed samples: 2841600 | consumed tokens: 5819596800 | elapsed time per iteration (s): 0.30 | learning rate: 1.071E-04 | global batch size: 256 | lm loss: 3.481216E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.059 | TFLOPs: 30.32 | 7: iteration 11110/ 21553 | consumed samples: 2844160 | consumed tokens: 5824839680 | elapsed time per iteration (s): 0.30 | learning rate: 1.070E-04 | global batch size: 256 | lm loss: 3.478028E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.795 | TFLOPs: 30.31 | 7: iteration 11120/ 21553 | consumed samples: 2846720 | consumed tokens: 5830082560 | elapsed time per iteration (s): 0.31 | learning rate: 1.069E-04 | global batch size: 256 | lm loss: 3.478998E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 836.044 | TFLOPs: 29.27 | 7: iteration 11130/ 21553 | consumed samples: 2849280 | consumed tokens: 5835325440 | elapsed time per iteration (s): 0.30 | learning rate: 1.067E-04 | global batch size: 256 | lm loss: 3.485382E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.034 | TFLOPs: 30.32 | 7: iteration 11140/ 21553 | consumed samples: 2851840 | consumed tokens: 5840568320 | elapsed time per iteration (s): 0.30 | learning rate: 1.066E-04 | global batch size: 256 | lm loss: 3.480761E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.484 | TFLOPs: 30.33 | 7: iteration 11150/ 21553 | consumed samples: 2854400 | consumed tokens: 5845811200 | elapsed time per iteration (s): 0.30 | learning rate: 1.065E-04 | global batch size: 256 | lm loss: 3.491139E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.178 | TFLOPs: 30.32 | 7: iteration 11160/ 21553 | consumed samples: 2856960 | consumed tokens: 5851054080 | elapsed time per iteration (s): 0.30 | learning rate: 1.063E-04 | global batch size: 256 | lm loss: 3.488515E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.582 | TFLOPs: 30.34 | 7: iteration 11170/ 21553 | consumed samples: 2859520 | consumed tokens: 5856296960 | elapsed time per iteration (s): 0.30 | learning rate: 1.062E-04 | global batch size: 256 | lm loss: 3.485668E+00 | grad norm: 0.456 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.873 | TFLOPs: 30.31 | 7: iteration 11180/ 21553 | consumed samples: 2862080 | consumed tokens: 5861539840 | elapsed time per iteration (s): 0.30 | learning rate: 1.061E-04 | global batch size: 256 | lm loss: 3.483554E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.910 | TFLOPs: 30.28 | 7: iteration 11190/ 21553 | consumed samples: 2864640 | consumed tokens: 5866782720 | elapsed time per iteration (s): 0.30 | learning rate: 1.060E-04 | global batch size: 256 | lm loss: 3.475839E+00 | grad norm: 0.456 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.287 | TFLOPs: 30.29 | 7: iteration 11200/ 21553 | consumed samples: 2867200 | consumed tokens: 5872025600 | elapsed time per iteration (s): 0.30 | learning rate: 1.058E-04 | global batch size: 256 | lm loss: 3.480733E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.383 | TFLOPs: 30.29 | 7: iteration 11210/ 21553 | consumed samples: 2869760 | consumed tokens: 5877268480 | elapsed time per iteration (s): 0.30 | learning rate: 1.057E-04 | global batch size: 256 | lm loss: 3.480773E+00 | grad norm: 0.393 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.148 | TFLOPs: 30.32 | 7: iteration 11220/ 21553 | consumed samples: 2872320 | consumed tokens: 5882511360 | elapsed time per iteration (s): 0.30 | learning rate: 1.056E-04 | global batch size: 256 | lm loss: 3.478417E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.366 | TFLOPs: 30.33 | 7: iteration 11230/ 21553 | consumed samples: 2874880 | consumed tokens: 5887754240 | elapsed time per iteration (s): 0.30 | learning rate: 1.054E-04 | global batch size: 256 | lm loss: 3.476939E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.261 | TFLOPs: 30.33 | 7: iteration 11240/ 21553 | consumed samples: 2877440 | consumed tokens: 5892997120 | elapsed time per iteration (s): 0.30 | learning rate: 1.053E-04 | global batch size: 256 | lm loss: 3.479892E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.538 | TFLOPs: 30.34 | 7: iteration 11250/ 21553 | consumed samples: 2880000 | consumed tokens: 5898240000 | elapsed time per iteration (s): 0.30 | learning rate: 1.052E-04 | global batch size: 256 | lm loss: 3.475893E+00 | grad norm: 0.449 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.707 | TFLOPs: 30.34 | 7: iteration 11260/ 21553 | consumed samples: 2882560 | consumed tokens: 5903482880 | elapsed time per iteration (s): 0.30 | learning rate: 1.050E-04 | global batch size: 256 | lm loss: 3.478431E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.276 | TFLOPs: 30.33 | 7: iteration 11270/ 21553 | consumed samples: 2885120 | consumed tokens: 5908725760 | elapsed time per iteration (s): 0.30 | learning rate: 1.049E-04 | global batch size: 256 | lm loss: 3.483459E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.186 | TFLOPs: 30.36 | 7: iteration 11280/ 21553 | consumed samples: 2887680 | consumed tokens: 5913968640 | elapsed time per iteration (s): 0.30 | learning rate: 1.048E-04 | global batch size: 256 | lm loss: 3.489210E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.555 | TFLOPs: 30.34 | 7: iteration 11290/ 21553 | consumed samples: 2890240 | consumed tokens: 5919211520 | elapsed time per iteration (s): 0.30 | learning rate: 1.046E-04 | global batch size: 256 | lm loss: 3.489042E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.656 | TFLOPs: 30.34 | 7: iteration 11300/ 21553 | consumed samples: 2892800 | consumed tokens: 5924454400 | elapsed time per iteration (s): 0.30 | learning rate: 1.045E-04 | global batch size: 256 | lm loss: 3.481274E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.393 | TFLOPs: 30.33 | 7: iteration 11310/ 21553 | consumed samples: 2895360 | consumed tokens: 5929697280 | elapsed time per iteration (s): 0.30 | learning rate: 1.044E-04 | global batch size: 256 | lm loss: 3.471778E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.730 | TFLOPs: 30.34 | 7: iteration 11320/ 21553 | consumed samples: 2897920 | consumed tokens: 5934940160 | elapsed time per iteration (s): 0.30 | learning rate: 1.042E-04 | global batch size: 256 | lm loss: 3.483895E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.521 | TFLOPs: 30.33 | 7: iteration 11330/ 21553 | consumed samples: 2900480 | consumed tokens: 5940183040 | elapsed time per iteration (s): 0.30 | learning rate: 1.041E-04 | global batch size: 256 | lm loss: 3.478096E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.285 | TFLOPs: 30.33 | 7: iteration 11340/ 21553 | consumed samples: 2903040 | consumed tokens: 5945425920 | elapsed time per iteration (s): 0.30 | learning rate: 1.040E-04 | global batch size: 256 | lm loss: 3.476935E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.895 | TFLOPs: 30.31 | 7: iteration 11350/ 21553 | consumed samples: 2905600 | consumed tokens: 5950668800 | elapsed time per iteration (s): 0.30 | learning rate: 1.038E-04 | global batch size: 256 | lm loss: 3.477710E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.065 | TFLOPs: 30.32 | 7: iteration 11360/ 21553 | consumed samples: 2908160 | consumed tokens: 5955911680 | elapsed time per iteration (s): 0.30 | learning rate: 1.037E-04 | global batch size: 256 | lm loss: 3.473494E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.769 | TFLOPs: 30.31 | 7: iteration 11370/ 21553 | consumed samples: 2910720 | consumed tokens: 5961154560 | elapsed time per iteration (s): 0.30 | learning rate: 1.036E-04 | global batch size: 256 | lm loss: 3.476336E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.847 | TFLOPs: 30.31 | 7: iteration 11380/ 21553 | consumed samples: 2913280 | consumed tokens: 5966397440 | elapsed time per iteration (s): 0.30 | learning rate: 1.034E-04 | global batch size: 256 | lm loss: 3.466453E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.771 | TFLOPs: 30.31 | 7: iteration 11390/ 21553 | consumed samples: 2915840 | consumed tokens: 5971640320 | elapsed time per iteration (s): 0.30 | learning rate: 1.033E-04 | global batch size: 256 | lm loss: 3.472049E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.623 | TFLOPs: 30.30 | 7: iteration 11400/ 21553 | consumed samples: 2918400 | consumed tokens: 5976883200 | elapsed time per iteration (s): 0.30 | learning rate: 1.032E-04 | global batch size: 256 | lm loss: 3.474117E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.111 | TFLOPs: 30.32 | 7: iteration 11410/ 21553 | consumed samples: 2920960 | consumed tokens: 5982126080 | elapsed time per iteration (s): 0.30 | learning rate: 1.030E-04 | global batch size: 256 | lm loss: 3.474647E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.015 | TFLOPs: 30.25 | 7: iteration 11420/ 21553 | consumed samples: 2923520 | consumed tokens: 5987368960 | elapsed time per iteration (s): 0.30 | learning rate: 1.029E-04 | global batch size: 256 | lm loss: 3.471074E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.785 | TFLOPs: 30.31 | 7: iteration 11430/ 21553 | consumed samples: 2926080 | consumed tokens: 5992611840 | elapsed time per iteration (s): 0.30 | learning rate: 1.028E-04 | global batch size: 256 | lm loss: 3.475668E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.642 | TFLOPs: 30.30 | 7: iteration 11440/ 21553 | consumed samples: 2928640 | consumed tokens: 5997854720 | elapsed time per iteration (s): 0.30 | learning rate: 1.026E-04 | global batch size: 256 | lm loss: 3.476446E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.339 | TFLOPs: 30.29 | 7: iteration 11450/ 21553 | consumed samples: 2931200 | consumed tokens: 6003097600 | elapsed time per iteration (s): 0.30 | learning rate: 1.025E-04 | global batch size: 256 | lm loss: 3.479917E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.688 | TFLOPs: 30.31 | 7: iteration 11460/ 21553 | consumed samples: 2933760 | consumed tokens: 6008340480 | elapsed time per iteration (s): 0.30 | learning rate: 1.024E-04 | global batch size: 256 | lm loss: 3.476341E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.442 | TFLOPs: 30.30 | 7: iteration 11470/ 21553 | consumed samples: 2936320 | consumed tokens: 6013583360 | elapsed time per iteration (s): 0.30 | learning rate: 1.022E-04 | global batch size: 256 | lm loss: 3.470049E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.609 | TFLOPs: 30.30 | 7: iteration 11480/ 21553 | consumed samples: 2938880 | consumed tokens: 6018826240 | elapsed time per iteration (s): 0.30 | learning rate: 1.021E-04 | global batch size: 256 | lm loss: 3.471270E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.439 | TFLOPs: 30.30 | 7: iteration 11490/ 21553 | consumed samples: 2941440 | consumed tokens: 6024069120 | elapsed time per iteration (s): 0.30 | learning rate: 1.020E-04 | global batch size: 256 | lm loss: 3.474570E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 849.771 | TFLOPs: 29.75 | 7: iteration 11500/ 21553 | consumed samples: 2944000 | consumed tokens: 6029312000 | elapsed time per iteration (s): 0.30 | learning rate: 1.019E-04 | global batch size: 256 | lm loss: 3.469035E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 840.643 | TFLOPs: 29.43 | 7: iteration 11510/ 21553 | consumed samples: 2946560 | consumed tokens: 6034554880 | elapsed time per iteration (s): 0.30 | learning rate: 1.017E-04 | global batch size: 256 | lm loss: 3.467540E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.920 | TFLOPs: 30.28 | 7: iteration 11520/ 21553 | consumed samples: 2949120 | consumed tokens: 6039797760 | elapsed time per iteration (s): 0.30 | learning rate: 1.016E-04 | global batch size: 256 | lm loss: 3.469558E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.328 | TFLOPs: 30.29 | 7: iteration 11530/ 21553 | consumed samples: 2951680 | consumed tokens: 6045040640 | elapsed time per iteration (s): 0.30 | learning rate: 1.015E-04 | global batch size: 256 | lm loss: 3.474889E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.446 | TFLOPs: 30.30 | 7: iteration 11540/ 21553 | consumed samples: 2954240 | consumed tokens: 6050283520 | elapsed time per iteration (s): 0.30 | learning rate: 1.013E-04 | global batch size: 256 | lm loss: 3.460758E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.601 | TFLOPs: 30.27 | 7: iteration 11550/ 21553 | consumed samples: 2956800 | consumed tokens: 6055526400 | elapsed time per iteration (s): 0.30 | learning rate: 1.012E-04 | global batch size: 256 | lm loss: 3.472905E+00 | grad norm: 0.472 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.049 | TFLOPs: 30.25 | 7: iteration 11560/ 21553 | consumed samples: 2959360 | consumed tokens: 6060769280 | elapsed time per iteration (s): 0.30 | learning rate: 1.011E-04 | global batch size: 256 | lm loss: 3.477110E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.012 | TFLOPs: 30.28 | 7: iteration 11570/ 21553 | consumed samples: 2961920 | consumed tokens: 6066012160 | elapsed time per iteration (s): 0.30 | learning rate: 1.009E-04 | global batch size: 256 | lm loss: 3.468501E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.657 | TFLOPs: 30.27 | 7: iteration 11580/ 21553 | consumed samples: 2964480 | consumed tokens: 6071255040 | elapsed time per iteration (s): 0.30 | learning rate: 1.008E-04 | global batch size: 256 | lm loss: 3.469975E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.427 | TFLOPs: 30.23 | 7: iteration 11590/ 21553 | consumed samples: 2967040 | consumed tokens: 6076497920 | elapsed time per iteration (s): 0.30 | learning rate: 1.007E-04 | global batch size: 256 | lm loss: 3.464762E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.914 | TFLOPs: 30.24 | 7: iteration 11600/ 21553 | consumed samples: 2969600 | consumed tokens: 6081740800 | elapsed time per iteration (s): 0.30 | learning rate: 1.005E-04 | global batch size: 256 | lm loss: 3.473625E+00 | grad norm: 0.477 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.552 | TFLOPs: 30.27 | 7: iteration 11610/ 21553 | consumed samples: 2972160 | consumed tokens: 6086983680 | elapsed time per iteration (s): 0.30 | learning rate: 1.004E-04 | global batch size: 256 | lm loss: 3.469656E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.055 | TFLOPs: 30.28 | 7: iteration 11620/ 21553 | consumed samples: 2974720 | consumed tokens: 6092226560 | elapsed time per iteration (s): 0.30 | learning rate: 1.003E-04 | global batch size: 256 | lm loss: 3.469639E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.804 | TFLOPs: 30.27 | 7: iteration 11630/ 21553 | consumed samples: 2977280 | consumed tokens: 6097469440 | elapsed time per iteration (s): 0.30 | learning rate: 1.001E-04 | global batch size: 256 | lm loss: 3.467694E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.069 | TFLOPs: 30.28 | 7: iteration 11640/ 21553 | consumed samples: 2979840 | consumed tokens: 6102712320 | elapsed time per iteration (s): 0.30 | learning rate: 1.000E-04 | global batch size: 256 | lm loss: 3.462105E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.586 | TFLOPs: 30.23 | 7: iteration 11650/ 21553 | consumed samples: 2982400 | consumed tokens: 6107955200 | elapsed time per iteration (s): 0.30 | learning rate: 9.987E-05 | global batch size: 256 | lm loss: 3.468061E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.671 | TFLOPs: 30.27 | 7: iteration 11660/ 21553 | consumed samples: 2984960 | consumed tokens: 6113198080 | elapsed time per iteration (s): 0.30 | learning rate: 9.974E-05 | global batch size: 256 | lm loss: 3.471376E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.511 | TFLOPs: 30.30 | 7: iteration 11670/ 21553 | consumed samples: 2987520 | consumed tokens: 6118440960 | elapsed time per iteration (s): 0.30 | learning rate: 9.961E-05 | global batch size: 256 | lm loss: 3.469095E+00 | grad norm: 0.460 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.983 | TFLOPs: 30.32 | 7: iteration 11680/ 21553 | consumed samples: 2990080 | consumed tokens: 6123683840 | elapsed time per iteration (s): 0.30 | learning rate: 9.948E-05 | global batch size: 256 | lm loss: 3.461526E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.788 | TFLOPs: 30.27 | 7: iteration 11690/ 21553 | consumed samples: 2992640 | consumed tokens: 6128926720 | elapsed time per iteration (s): 0.30 | learning rate: 9.935E-05 | global batch size: 256 | lm loss: 3.460766E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.737 | TFLOPs: 30.31 | 7: iteration 11700/ 21553 | consumed samples: 2995200 | consumed tokens: 6134169600 | elapsed time per iteration (s): 0.30 | learning rate: 9.922E-05 | global batch size: 256 | lm loss: 3.470131E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.469 | TFLOPs: 30.26 | 7: iteration 11710/ 21553 | consumed samples: 2997760 | consumed tokens: 6139412480 | elapsed time per iteration (s): 0.30 | learning rate: 9.909E-05 | global batch size: 256 | lm loss: 3.470738E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.439 | TFLOPs: 30.30 | 7: iteration 11720/ 21553 | consumed samples: 3000320 | consumed tokens: 6144655360 | elapsed time per iteration (s): 0.30 | learning rate: 9.895E-05 | global batch size: 256 | lm loss: 3.467196E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.463 | TFLOPs: 30.30 | 7: iteration 11730/ 21553 | consumed samples: 3002880 | consumed tokens: 6149898240 | elapsed time per iteration (s): 0.30 | learning rate: 9.882E-05 | global batch size: 256 | lm loss: 3.463244E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 842.192 | TFLOPs: 29.48 | 7: iteration 11740/ 21553 | consumed samples: 3005440 | consumed tokens: 6155141120 | elapsed time per iteration (s): 0.30 | learning rate: 9.869E-05 | global batch size: 256 | lm loss: 3.465920E+00 | grad norm: 0.393 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.512 | TFLOPs: 30.30 | 7: iteration 11750/ 21553 | consumed samples: 3008000 | consumed tokens: 6160384000 | elapsed time per iteration (s): 0.30 | learning rate: 9.856E-05 | global batch size: 256 | lm loss: 3.460855E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.985 | TFLOPs: 30.32 | 7: iteration 11760/ 21553 | consumed samples: 3010560 | consumed tokens: 6165626880 | elapsed time per iteration (s): 0.30 | learning rate: 9.843E-05 | global batch size: 256 | lm loss: 3.462743E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.758 | TFLOPs: 30.31 | 7: iteration 11770/ 21553 | consumed samples: 3013120 | consumed tokens: 6170869760 | elapsed time per iteration (s): 0.30 | learning rate: 9.830E-05 | global batch size: 256 | lm loss: 3.461045E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.761 | TFLOPs: 30.31 | 7: iteration 11780/ 21553 | consumed samples: 3015680 | consumed tokens: 6176112640 | elapsed time per iteration (s): 0.30 | learning rate: 9.817E-05 | global batch size: 256 | lm loss: 3.472029E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.841 | TFLOPs: 30.35 | 7: iteration 11790/ 21553 | consumed samples: 3018240 | consumed tokens: 6181355520 | elapsed time per iteration (s): 0.30 | learning rate: 9.803E-05 | global batch size: 256 | lm loss: 3.464385E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.420 | TFLOPs: 30.33 | 7: iteration 11800/ 21553 | consumed samples: 3020800 | consumed tokens: 6186598400 | elapsed time per iteration (s): 0.30 | learning rate: 9.790E-05 | global batch size: 256 | lm loss: 3.467997E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.139 | TFLOPs: 30.32 | 7: iteration 11810/ 21553 | consumed samples: 3023360 | consumed tokens: 6191841280 | elapsed time per iteration (s): 0.30 | learning rate: 9.777E-05 | global batch size: 256 | lm loss: 3.465111E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.283 | TFLOPs: 30.29 | 7: iteration 11820/ 21553 | consumed samples: 3025920 | consumed tokens: 6197084160 | elapsed time per iteration (s): 0.30 | learning rate: 9.764E-05 | global batch size: 256 | lm loss: 3.466401E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.331 | TFLOPs: 30.26 | 7: iteration 11830/ 21553 | consumed samples: 3028480 | consumed tokens: 6202327040 | elapsed time per iteration (s): 0.30 | learning rate: 9.751E-05 | global batch size: 256 | lm loss: 3.461584E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.736 | TFLOPs: 30.24 | 7: iteration 11840/ 21553 | consumed samples: 3031040 | consumed tokens: 6207569920 | elapsed time per iteration (s): 0.30 | learning rate: 9.738E-05 | global batch size: 256 | lm loss: 3.468016E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.410 | TFLOPs: 30.26 | 7: iteration 11850/ 21553 | consumed samples: 3033600 | consumed tokens: 6212812800 | elapsed time per iteration (s): 0.30 | learning rate: 9.725E-05 | global batch size: 256 | lm loss: 3.457066E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.689 | TFLOPs: 30.24 | 7: iteration 11860/ 21553 | consumed samples: 3036160 | consumed tokens: 6218055680 | elapsed time per iteration (s): 0.30 | learning rate: 9.712E-05 | global batch size: 256 | lm loss: 3.461216E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.310 | TFLOPs: 30.22 | 7: iteration 11870/ 21553 | consumed samples: 3038720 | consumed tokens: 6223298560 | elapsed time per iteration (s): 0.30 | learning rate: 9.698E-05 | global batch size: 256 | lm loss: 3.465809E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.005 | TFLOPs: 30.25 | 7: iteration 11880/ 21553 | consumed samples: 3041280 | consumed tokens: 6228541440 | elapsed time per iteration (s): 0.30 | learning rate: 9.685E-05 | global batch size: 256 | lm loss: 3.466081E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.828 | TFLOPs: 30.28 | 7: iteration 11890/ 21553 | consumed samples: 3043840 | consumed tokens: 6233784320 | elapsed time per iteration (s): 0.30 | learning rate: 9.672E-05 | global batch size: 256 | lm loss: 3.453565E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.219 | TFLOPs: 30.29 | 7: iteration 11900/ 21553 | consumed samples: 3046400 | consumed tokens: 6239027200 | elapsed time per iteration (s): 0.30 | learning rate: 9.659E-05 | global batch size: 256 | lm loss: 3.465266E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.218 | TFLOPs: 30.29 | 7: iteration 11910/ 21553 | consumed samples: 3048960 | consumed tokens: 6244270080 | elapsed time per iteration (s): 0.30 | learning rate: 9.646E-05 | global batch size: 256 | lm loss: 3.459411E+00 | grad norm: 0.480 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.684 | TFLOPs: 30.31 | 7: iteration 11920/ 21553 | consumed samples: 3051520 | consumed tokens: 6249512960 | elapsed time per iteration (s): 0.30 | learning rate: 9.633E-05 | global batch size: 256 | lm loss: 3.465755E+00 | grad norm: 0.464 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.886 | TFLOPs: 30.31 | 7: iteration 11930/ 21553 | consumed samples: 3054080 | consumed tokens: 6254755840 | elapsed time per iteration (s): 0.30 | learning rate: 9.620E-05 | global batch size: 256 | lm loss: 3.464436E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.973 | TFLOPs: 30.32 | 7: iteration 11940/ 21553 | consumed samples: 3056640 | consumed tokens: 6259998720 | elapsed time per iteration (s): 0.30 | learning rate: 9.607E-05 | global batch size: 256 | lm loss: 3.465108E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.359 | TFLOPs: 30.29 | 7: iteration 11950/ 21553 | consumed samples: 3059200 | consumed tokens: 6265241600 | elapsed time per iteration (s): 0.30 | learning rate: 9.594E-05 | global batch size: 256 | lm loss: 3.462003E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.302 | TFLOPs: 30.26 | 7: iteration 11960/ 21553 | consumed samples: 3061760 | consumed tokens: 6270484480 | elapsed time per iteration (s): 0.30 | learning rate: 9.581E-05 | global batch size: 256 | lm loss: 3.449902E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.150 | TFLOPs: 30.25 | 7: iteration 11970/ 21553 | consumed samples: 3064320 | consumed tokens: 6275727360 | elapsed time per iteration (s): 0.30 | learning rate: 9.567E-05 | global batch size: 256 | lm loss: 3.459981E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.523 | TFLOPs: 30.26 | 7: iteration 11980/ 21553 | consumed samples: 3066880 | consumed tokens: 6280970240 | elapsed time per iteration (s): 0.30 | learning rate: 9.554E-05 | global batch size: 256 | lm loss: 3.456469E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.083 | TFLOPs: 30.28 | 7: iteration 11990/ 21553 | consumed samples: 3069440 | consumed tokens: 6286213120 | elapsed time per iteration (s): 0.30 | learning rate: 9.541E-05 | global batch size: 256 | lm loss: 3.455710E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.439 | TFLOPs: 30.30 | 0: [2023-03-17 00:55:19,603] [INFO] [logging.py:68:log_dist] [Rank 0] step=12000, skipped=0, lr=[9.528206376265585e-05, 9.528206376265585e-05, 9.528206376265585e-05], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 12000/ 21553 | consumed samples: 3072000 | consumed tokens: 6291456000 | elapsed time per iteration (s): 0.30 | learning rate: 9.528E-05 | global batch size: 256 | lm loss: 3.450189E+00 | grad norm: 0.389 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.890 | TFLOPs: 30.31 | 0: steps: 12000 loss: 3.4635 iter time (s): 0.294 samples/sec: 870.532 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 12000 | lm loss value: 3.901426E+00 | lm loss PPL: 4.947294E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 12000 to checkpoints_146m14b100mdedup 0: [2023-03-17 00:55:19,724] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step12000 is begin to save! 0: [2023-03-17 00:55:19,727] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_01-model_00-model_states.pt... 0: [2023-03-17 00:55:19,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_01-model_00-model_states.pt. 0: [2023-03-17 00:55:19,813] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_03-model_00-model_states.pt... 0: [2023-03-17 00:55:19,829] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_03-model_00-model_states.pt. 0: [2023-03-17 00:55:19,829] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_04-model_00-model_states.pt... 0: [2023-03-17 00:55:19,844] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_04-model_00-model_states.pt. 0: [2023-03-17 00:55:19,844] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_05-model_00-model_states.pt... 0: [2023-03-17 00:55:19,860] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_05-model_00-model_states.pt. 0: [2023-03-17 00:55:19,860] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_06-model_00-model_states.pt... 0: [2023-03-17 00:55:19,874] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_06-model_00-model_states.pt. 0: [2023-03-17 00:55:19,875] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_07-model_00-model_states.pt... 0: [2023-03-17 00:55:19,889] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_07-model_00-model_states.pt. 0: [2023-03-17 00:55:19,890] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_08-model_00-model_states.pt... 0: [2023-03-17 00:55:19,904] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_08-model_00-model_states.pt. 0: [2023-03-17 00:55:19,905] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_09-model_00-model_states.pt... 0: [2023-03-17 00:55:19,919] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_09-model_00-model_states.pt. 0: [2023-03-17 00:55:19,919] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_10-model_00-model_states.pt... 0: [2023-03-17 00:55:19,934] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_10-model_00-model_states.pt. 0: [2023-03-17 00:55:19,934] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_11-model_00-model_states.pt... 0: [2023-03-17 00:55:19,949] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_11-model_00-model_states.pt. 0: [2023-03-17 00:55:19,949] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_12-model_00-model_states.pt... 0: [2023-03-17 00:55:19,964] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_12-model_00-model_states.pt. 0: [2023-03-17 00:55:19,964] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_13-model_00-model_states.pt... 0: [2023-03-17 00:55:19,979] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_13-model_00-model_states.pt. 0: [2023-03-17 00:55:19,979] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_14-model_00-model_states.pt... 0: [2023-03-17 00:55:19,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_14-model_00-model_states.pt. 0: [2023-03-17 00:55:19,994] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_15-model_00-model_states.pt... 0: [2023-03-17 00:55:20,009] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_15-model_00-model_states.pt. 0: [2023-03-17 00:55:20,009] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_16-model_00-model_states.pt... 0: [2023-03-17 00:55:20,023] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_16-model_00-model_states.pt. 0: [2023-03-17 00:55:20,024] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_17-model_00-model_states.pt... 0: [2023-03-17 00:55:20,038] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_17-model_00-model_states.pt. 0: [2023-03-17 00:55:20,038] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/layer_19-model_00-model_states.pt... 0: [2023-03-17 00:55:20,039] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/layer_19-model_00-model_states.pt. 0: [2023-03-17 00:55:20,040] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step12000/mp_rank_00_model_states.pt 0: [2023-03-17 00:55:20,040] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/mp_rank_00_model_states.pt... 0: [2023-03-17 00:55:20,042] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/mp_rank_00_model_states.pt. 0: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 6: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 5: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 2: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 7: [2023-03-17 00:55:20,059] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 0: [2023-03-17 00:55:20,090] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:55:20,090] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:55:20,090] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 00:55:20,090] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 3: [2023-03-17 00:55:20,091] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:55:20,091] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 00:55:20,091] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 7: [2023-03-17 00:55:20,092] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:55:20,092] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 00:55:20,092] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 4: [2023-03-17 00:55:20,092] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:55:20,093] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 5: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:55:20,093] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 6: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:55:20,093] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 00:55:20,093] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,093] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 1: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:55:20,093] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 5: [2023-03-17 00:55:20,093] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 1: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 5: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 4: [2023-03-17 00:55:20,093] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:55:20,094] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 5: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:55:20,094] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 2: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:55:20,094] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 0: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:55:20,095] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:55:20,095] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 0: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:55:20,095] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 3: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:55:20,096] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 00:55:20,096] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 3: [2023-03-17 00:55:20,096] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:55:20,096] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 00:55:20,096] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:55:20,097] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 1: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 1: [2023-03-17 00:55:20,097] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 7: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:55:20,095] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 7: [2023-03-17 00:55:20,095] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:55:20,096] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 00:55:20,096] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,098] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 7: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:55:20,097] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,098] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 7: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:55:20,097] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 3: [2023-03-17 00:55:20,098] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:55:20,098] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 00:55:20,098] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 3: [2023-03-17 00:55:20,098] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:55:20,098] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 00:55:20,098] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:55:20,098] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 3: [2023-03-17 00:55:20,098] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 00:55:20,098] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:55:20,098] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 3: [2023-03-17 00:55:20,098] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 1: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:55:20,099] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 1: [2023-03-17 00:55:20,099] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 00:55:20,099] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 00:55:20,099] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 1: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 1: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 5: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:55:20,100] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 5: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:55:20,100] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 0: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:55:20,100] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 00:55:20,100] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 0: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 0: [2023-03-17 00:55:20,100] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 0: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 00:55:20,100] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 00:55:20,101] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 4: [2023-03-17 00:55:20,100] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:55:20,101] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 00:55:20,101] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 4: [2023-03-17 00:55:20,101] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:55:20,101] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:55:20,101] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 00:55:20,101] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 00:55:20,101] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 4: [2023-03-17 00:55:20,101] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 2: [2023-03-17 00:55:20,094] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 2: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:55:20,094] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 2: [2023-03-17 00:55:20,094] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:55:20,095] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 2: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:55:20,095] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 00:55:20,095] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 2: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:55:20,097] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 2: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:55:20,097] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 00:55:20,097] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 2: [2023-03-17 00:55:20,097] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 2: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 00:55:20,099] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 00:55:20,099] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 1: [2023-03-17 00:55:20,103] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 00:55:20,103] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 00:55:20,103] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,103] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:55:20,103] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 00:55:20,103] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 6: [2023-03-17 00:55:20,103] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 00:55:20,104] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 00:55:20,104] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 4: [2023-03-17 00:55:20,104] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:55:20,104] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 00:55:20,104] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 4: [2023-03-17 00:55:20,104] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:55:20,104] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 00:55:20,105] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 5: [2023-03-17 00:55:20,105] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:55:20,105] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 00:55:20,105] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 5: [2023-03-17 00:55:20,105] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 00:55:20,105] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 00:55:20,105] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 4: [2023-03-17 00:55:20,105] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:55:20,105] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 5: [2023-03-17 00:55:20,105] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 4: [2023-03-17 00:55:20,105] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 5: [2023-03-17 00:55:20,105] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 00:55:20,105] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 3: [2023-03-17 00:55:20,107] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 00:55:20,107] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 00:55:20,107] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 7: [2023-03-17 00:55:20,108] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:55:20,108] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 00:55:20,108] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 7: [2023-03-17 00:55:20,108] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 00:55:20,108] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 00:55:20,108] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 0: [2023-03-17 00:55:20,111] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step12000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 00:55:20,111] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step12000 is ready now! 0: successfully saved checkpoint at iteration 12000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 401.98 7: iteration 12010/ 21553 | consumed samples: 3074560 | consumed tokens: 6296698880 | elapsed time per iteration (s): 0.35 | learning rate: 9.515E-05 | global batch size: 256 | lm loss: 3.461000E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 738.000 | TFLOPs: 25.84 | 7: iteration 12020/ 21553 | consumed samples: 3077120 | consumed tokens: 6301941760 | elapsed time per iteration (s): 0.30 | learning rate: 9.502E-05 | global batch size: 256 | lm loss: 3.456076E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.551 | TFLOPs: 30.27 | 7: iteration 12030/ 21553 | consumed samples: 3079680 | consumed tokens: 6307184640 | elapsed time per iteration (s): 0.30 | learning rate: 9.489E-05 | global batch size: 256 | lm loss: 3.452217E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.608 | TFLOPs: 30.23 | 7: iteration 12040/ 21553 | consumed samples: 3082240 | consumed tokens: 6312427520 | elapsed time per iteration (s): 0.30 | learning rate: 9.476E-05 | global batch size: 256 | lm loss: 3.465364E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.513 | TFLOPs: 30.26 | 7: iteration 12050/ 21553 | consumed samples: 3084800 | consumed tokens: 6317670400 | elapsed time per iteration (s): 0.30 | learning rate: 9.463E-05 | global batch size: 256 | lm loss: 3.462815E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.179 | TFLOPs: 30.25 | 7: iteration 12060/ 21553 | consumed samples: 3087360 | consumed tokens: 6322913280 | elapsed time per iteration (s): 0.30 | learning rate: 9.450E-05 | global batch size: 256 | lm loss: 3.460719E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.699 | TFLOPs: 30.27 | 7: iteration 12070/ 21553 | consumed samples: 3089920 | consumed tokens: 6328156160 | elapsed time per iteration (s): 0.30 | learning rate: 9.437E-05 | global batch size: 256 | lm loss: 3.460153E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.441 | TFLOPs: 30.26 | 7: iteration 12080/ 21553 | consumed samples: 3092480 | consumed tokens: 6333399040 | elapsed time per iteration (s): 0.30 | learning rate: 9.424E-05 | global batch size: 256 | lm loss: 3.456115E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.387 | TFLOPs: 30.26 | 7: iteration 12090/ 21553 | consumed samples: 3095040 | consumed tokens: 6338641920 | elapsed time per iteration (s): 0.30 | learning rate: 9.411E-05 | global batch size: 256 | lm loss: 3.453911E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.792 | TFLOPs: 30.27 | 7: iteration 12100/ 21553 | consumed samples: 3097600 | consumed tokens: 6343884800 | elapsed time per iteration (s): 0.30 | learning rate: 9.398E-05 | global batch size: 256 | lm loss: 3.450656E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.017 | TFLOPs: 30.18 | 7: iteration 12110/ 21553 | consumed samples: 3100160 | consumed tokens: 6349127680 | elapsed time per iteration (s): 0.30 | learning rate: 9.385E-05 | global batch size: 256 | lm loss: 3.458420E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.177 | TFLOPs: 30.18 | 7: iteration 12120/ 21553 | consumed samples: 3102720 | consumed tokens: 6354370560 | elapsed time per iteration (s): 0.30 | learning rate: 9.372E-05 | global batch size: 256 | lm loss: 3.456826E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.377 | TFLOPs: 30.22 | 7: iteration 12130/ 21553 | consumed samples: 3105280 | consumed tokens: 6359613440 | elapsed time per iteration (s): 0.30 | learning rate: 9.359E-05 | global batch size: 256 | lm loss: 3.451901E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.272 | TFLOPs: 30.29 | 7: iteration 12140/ 21553 | consumed samples: 3107840 | consumed tokens: 6364856320 | elapsed time per iteration (s): 0.30 | learning rate: 9.346E-05 | global batch size: 256 | lm loss: 3.456087E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.688 | TFLOPs: 30.27 | 7: iteration 12150/ 21553 | consumed samples: 3110400 | consumed tokens: 6370099200 | elapsed time per iteration (s): 0.30 | learning rate: 9.332E-05 | global batch size: 256 | lm loss: 3.445337E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.427 | TFLOPs: 30.23 | 7: iteration 12160/ 21553 | consumed samples: 3112960 | consumed tokens: 6375342080 | elapsed time per iteration (s): 0.30 | learning rate: 9.319E-05 | global batch size: 256 | lm loss: 3.451311E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.308 | TFLOPs: 30.26 | 7: iteration 12170/ 21553 | consumed samples: 3115520 | consumed tokens: 6380584960 | elapsed time per iteration (s): 0.30 | learning rate: 9.306E-05 | global batch size: 256 | lm loss: 3.449361E+00 | grad norm: 0.454 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.700 | TFLOPs: 30.27 | 7: iteration 12180/ 21553 | consumed samples: 3118080 | consumed tokens: 6385827840 | elapsed time per iteration (s): 0.30 | learning rate: 9.293E-05 | global batch size: 256 | lm loss: 3.452225E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.675 | TFLOPs: 30.27 | 7: iteration 12190/ 21553 | consumed samples: 3120640 | consumed tokens: 6391070720 | elapsed time per iteration (s): 0.30 | learning rate: 9.280E-05 | global batch size: 256 | lm loss: 3.448108E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.094 | TFLOPs: 30.28 | 7: iteration 12200/ 21553 | consumed samples: 3123200 | consumed tokens: 6396313600 | elapsed time per iteration (s): 0.30 | learning rate: 9.267E-05 | global batch size: 256 | lm loss: 3.447927E+00 | grad norm: 0.458 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.690 | TFLOPs: 30.31 | 7: iteration 12210/ 21553 | consumed samples: 3125760 | consumed tokens: 6401556480 | elapsed time per iteration (s): 0.30 | learning rate: 9.254E-05 | global batch size: 256 | lm loss: 3.455567E+00 | grad norm: 0.466 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.051 | TFLOPs: 30.28 | 7: iteration 12220/ 21553 | consumed samples: 3128320 | consumed tokens: 6406799360 | elapsed time per iteration (s): 0.30 | learning rate: 9.241E-05 | global batch size: 256 | lm loss: 3.454405E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.139 | TFLOPs: 30.18 | 7: iteration 12230/ 21553 | consumed samples: 3130880 | consumed tokens: 6412042240 | elapsed time per iteration (s): 0.30 | learning rate: 9.228E-05 | global batch size: 256 | lm loss: 3.448365E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.090 | TFLOPs: 30.25 | 7: iteration 12240/ 21553 | consumed samples: 3133440 | consumed tokens: 6417285120 | elapsed time per iteration (s): 0.30 | learning rate: 9.215E-05 | global batch size: 256 | lm loss: 3.454311E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.355 | TFLOPs: 30.29 | 7: iteration 12250/ 21553 | consumed samples: 3136000 | consumed tokens: 6422528000 | elapsed time per iteration (s): 0.30 | learning rate: 9.202E-05 | global batch size: 256 | lm loss: 3.452496E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.063 | TFLOPs: 30.28 | 7: iteration 12260/ 21553 | consumed samples: 3138560 | consumed tokens: 6427770880 | elapsed time per iteration (s): 0.30 | learning rate: 9.189E-05 | global batch size: 256 | lm loss: 3.442028E+00 | grad norm: 0.394 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.412 | TFLOPs: 30.26 | 7: iteration 12270/ 21553 | consumed samples: 3141120 | consumed tokens: 6433013760 | elapsed time per iteration (s): 0.30 | learning rate: 9.177E-05 | global batch size: 256 | lm loss: 3.446137E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 846.189 | TFLOPs: 29.62 | 7: iteration 12280/ 21553 | consumed samples: 3143680 | consumed tokens: 6438256640 | elapsed time per iteration (s): 0.30 | learning rate: 9.164E-05 | global batch size: 256 | lm loss: 3.451467E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.321 | TFLOPs: 30.29 | 7: iteration 12290/ 21553 | consumed samples: 3146240 | consumed tokens: 6443499520 | elapsed time per iteration (s): 0.30 | learning rate: 9.151E-05 | global batch size: 256 | lm loss: 3.451403E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.894 | TFLOPs: 30.28 | 7: iteration 12300/ 21553 | consumed samples: 3148800 | consumed tokens: 6448742400 | elapsed time per iteration (s): 0.30 | learning rate: 9.138E-05 | global batch size: 256 | lm loss: 3.454409E+00 | grad norm: 0.455 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.563 | TFLOPs: 30.27 | 7: iteration 12310/ 21553 | consumed samples: 3151360 | consumed tokens: 6453985280 | elapsed time per iteration (s): 0.30 | learning rate: 9.125E-05 | global batch size: 256 | lm loss: 3.456109E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.386 | TFLOPs: 30.22 | 7: iteration 12320/ 21553 | consumed samples: 3153920 | consumed tokens: 6459228160 | elapsed time per iteration (s): 0.30 | learning rate: 9.112E-05 | global batch size: 256 | lm loss: 3.447805E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.878 | TFLOPs: 30.24 | 7: iteration 12330/ 21553 | consumed samples: 3156480 | consumed tokens: 6464471040 | elapsed time per iteration (s): 0.30 | learning rate: 9.099E-05 | global batch size: 256 | lm loss: 3.447916E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.629 | TFLOPs: 30.27 | 7: iteration 12340/ 21553 | consumed samples: 3159040 | consumed tokens: 6469713920 | elapsed time per iteration (s): 0.30 | learning rate: 9.086E-05 | global batch size: 256 | lm loss: 3.446889E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.810 | TFLOPs: 30.27 | 7: iteration 12350/ 21553 | consumed samples: 3161600 | consumed tokens: 6474956800 | elapsed time per iteration (s): 0.30 | learning rate: 9.073E-05 | global batch size: 256 | lm loss: 3.445940E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.449 | TFLOPs: 30.30 | 7: iteration 12360/ 21553 | consumed samples: 3164160 | consumed tokens: 6480199680 | elapsed time per iteration (s): 0.30 | learning rate: 9.060E-05 | global batch size: 256 | lm loss: 3.441375E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.822 | TFLOPs: 30.28 | 7: iteration 12370/ 21553 | consumed samples: 3166720 | consumed tokens: 6485442560 | elapsed time per iteration (s): 0.30 | learning rate: 9.047E-05 | global batch size: 256 | lm loss: 3.443521E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.848 | TFLOPs: 30.28 | 7: iteration 12380/ 21553 | consumed samples: 3169280 | consumed tokens: 6490685440 | elapsed time per iteration (s): 0.30 | learning rate: 9.034E-05 | global batch size: 256 | lm loss: 3.449087E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.905 | TFLOPs: 30.28 | 7: iteration 12390/ 21553 | consumed samples: 3171840 | consumed tokens: 6495928320 | elapsed time per iteration (s): 0.30 | learning rate: 9.021E-05 | global batch size: 256 | lm loss: 3.450904E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.524 | TFLOPs: 30.26 | 7: iteration 12400/ 21553 | consumed samples: 3174400 | consumed tokens: 6501171200 | elapsed time per iteration (s): 0.30 | learning rate: 9.008E-05 | global batch size: 256 | lm loss: 3.448396E+00 | grad norm: 0.451 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.676 | TFLOPs: 30.27 | 7: iteration 12410/ 21553 | consumed samples: 3176960 | consumed tokens: 6506414080 | elapsed time per iteration (s): 0.30 | learning rate: 8.995E-05 | global batch size: 256 | lm loss: 3.446156E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.610 | TFLOPs: 30.27 | 7: iteration 12420/ 21553 | consumed samples: 3179520 | consumed tokens: 6511656960 | elapsed time per iteration (s): 0.30 | learning rate: 8.982E-05 | global batch size: 256 | lm loss: 3.444762E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.457 | TFLOPs: 30.26 | 7: iteration 12430/ 21553 | consumed samples: 3182080 | consumed tokens: 6516899840 | elapsed time per iteration (s): 0.30 | learning rate: 8.969E-05 | global batch size: 256 | lm loss: 3.441819E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.337 | TFLOPs: 30.29 | 7: iteration 12440/ 21553 | consumed samples: 3184640 | consumed tokens: 6522142720 | elapsed time per iteration (s): 0.30 | learning rate: 8.957E-05 | global batch size: 256 | lm loss: 3.443051E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.197 | TFLOPs: 30.29 | 7: iteration 12450/ 21553 | consumed samples: 3187200 | consumed tokens: 6527385600 | elapsed time per iteration (s): 0.30 | learning rate: 8.944E-05 | global batch size: 256 | lm loss: 3.445733E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.357 | TFLOPs: 30.29 | 7: iteration 12460/ 21553 | consumed samples: 3189760 | consumed tokens: 6532628480 | elapsed time per iteration (s): 0.30 | learning rate: 8.931E-05 | global batch size: 256 | lm loss: 3.447690E+00 | grad norm: 0.451 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.381 | TFLOPs: 30.29 | 7: iteration 12470/ 21553 | consumed samples: 3192320 | consumed tokens: 6537871360 | elapsed time per iteration (s): 0.30 | learning rate: 8.918E-05 | global batch size: 256 | lm loss: 3.441108E+00 | grad norm: 0.395 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.350 | TFLOPs: 30.29 | 7: iteration 12480/ 21553 | consumed samples: 3194880 | consumed tokens: 6543114240 | elapsed time per iteration (s): 0.30 | learning rate: 8.905E-05 | global batch size: 256 | lm loss: 3.457177E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.157 | TFLOPs: 30.29 | 7: iteration 12490/ 21553 | consumed samples: 3197440 | consumed tokens: 6548357120 | elapsed time per iteration (s): 0.30 | learning rate: 8.892E-05 | global batch size: 256 | lm loss: 3.444442E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.922 | TFLOPs: 30.28 | 7: iteration 12500/ 21553 | consumed samples: 3200000 | consumed tokens: 6553600000 | elapsed time per iteration (s): 0.30 | learning rate: 8.879E-05 | global batch size: 256 | lm loss: 3.446003E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.433 | TFLOPs: 30.16 | 7: iteration 12510/ 21553 | consumed samples: 3202560 | consumed tokens: 6558842880 | elapsed time per iteration (s): 0.30 | learning rate: 8.866E-05 | global batch size: 256 | lm loss: 3.449153E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.085 | TFLOPs: 30.28 | 7: iteration 12520/ 21553 | consumed samples: 3205120 | consumed tokens: 6564085760 | elapsed time per iteration (s): 0.30 | learning rate: 8.853E-05 | global batch size: 256 | lm loss: 3.450100E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.241 | TFLOPs: 30.29 | 7: iteration 12530/ 21553 | consumed samples: 3207680 | consumed tokens: 6569328640 | elapsed time per iteration (s): 0.30 | learning rate: 8.841E-05 | global batch size: 256 | lm loss: 3.447485E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.983 | TFLOPs: 30.28 | 7: iteration 12540/ 21553 | consumed samples: 3210240 | consumed tokens: 6574571520 | elapsed time per iteration (s): 0.30 | learning rate: 8.828E-05 | global batch size: 256 | lm loss: 3.449395E+00 | grad norm: 0.460 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.378 | TFLOPs: 30.29 | 7: iteration 12550/ 21553 | consumed samples: 3212800 | consumed tokens: 6579814400 | elapsed time per iteration (s): 0.30 | learning rate: 8.815E-05 | global batch size: 256 | lm loss: 3.442807E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.242 | TFLOPs: 30.29 | 7: iteration 12560/ 21553 | consumed samples: 3215360 | consumed tokens: 6585057280 | elapsed time per iteration (s): 0.30 | learning rate: 8.802E-05 | global batch size: 256 | lm loss: 3.442651E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.764 | TFLOPs: 30.03 | 7: iteration 12570/ 21553 | consumed samples: 3217920 | consumed tokens: 6590300160 | elapsed time per iteration (s): 0.31 | learning rate: 8.789E-05 | global batch size: 256 | lm loss: 3.445271E+00 | grad norm: 0.397 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 818.012 | TFLOPs: 28.64 | 7: iteration 12580/ 21553 | consumed samples: 3220480 | consumed tokens: 6595543040 | elapsed time per iteration (s): 0.33 | learning rate: 8.776E-05 | global batch size: 256 | lm loss: 3.441697E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 787.601 | TFLOPs: 27.57 | 7: iteration 12590/ 21553 | consumed samples: 3223040 | consumed tokens: 6600785920 | elapsed time per iteration (s): 0.33 | learning rate: 8.763E-05 | global batch size: 256 | lm loss: 3.445213E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 768.462 | TFLOPs: 26.90 | 7: iteration 12600/ 21553 | consumed samples: 3225600 | consumed tokens: 6606028800 | elapsed time per iteration (s): 0.30 | learning rate: 8.751E-05 | global batch size: 256 | lm loss: 3.450994E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.109 | TFLOPs: 30.01 | 7: iteration 12610/ 21553 | consumed samples: 3228160 | consumed tokens: 6611271680 | elapsed time per iteration (s): 0.30 | learning rate: 8.738E-05 | global batch size: 256 | lm loss: 3.439100E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 850.568 | TFLOPs: 29.78 | 7: iteration 12620/ 21553 | consumed samples: 3230720 | consumed tokens: 6616514560 | elapsed time per iteration (s): 0.30 | learning rate: 8.725E-05 | global batch size: 256 | lm loss: 3.441901E+00 | grad norm: 0.449 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 842.142 | TFLOPs: 29.48 | 7: iteration 12630/ 21553 | consumed samples: 3233280 | consumed tokens: 6621757440 | elapsed time per iteration (s): 0.30 | learning rate: 8.712E-05 | global batch size: 256 | lm loss: 3.436263E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 853.143 | TFLOPs: 29.87 | 7: iteration 12640/ 21553 | consumed samples: 3235840 | consumed tokens: 6627000320 | elapsed time per iteration (s): 0.30 | learning rate: 8.699E-05 | global batch size: 256 | lm loss: 3.446952E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.579 | TFLOPs: 30.20 | 7: iteration 12650/ 21553 | consumed samples: 3238400 | consumed tokens: 6632243200 | elapsed time per iteration (s): 0.30 | learning rate: 8.687E-05 | global batch size: 256 | lm loss: 3.434644E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.190 | TFLOPs: 30.22 | 7: iteration 12660/ 21553 | consumed samples: 3240960 | consumed tokens: 6637486080 | elapsed time per iteration (s): 0.30 | learning rate: 8.674E-05 | global batch size: 256 | lm loss: 3.437071E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.772 | TFLOPs: 30.03 | 7: iteration 12670/ 21553 | consumed samples: 3243520 | consumed tokens: 6642728960 | elapsed time per iteration (s): 0.30 | learning rate: 8.661E-05 | global batch size: 256 | lm loss: 3.442380E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.315 | TFLOPs: 30.22 | 7: iteration 12680/ 21553 | consumed samples: 3246080 | consumed tokens: 6647971840 | elapsed time per iteration (s): 0.30 | learning rate: 8.648E-05 | global batch size: 256 | lm loss: 3.437535E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.071 | TFLOPs: 30.21 | 7: iteration 12690/ 21553 | consumed samples: 3248640 | consumed tokens: 6653214720 | elapsed time per iteration (s): 0.30 | learning rate: 8.635E-05 | global batch size: 256 | lm loss: 3.442212E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.008 | TFLOPs: 30.18 | 7: iteration 12700/ 21553 | consumed samples: 3251200 | consumed tokens: 6658457600 | elapsed time per iteration (s): 0.30 | learning rate: 8.623E-05 | global batch size: 256 | lm loss: 3.444978E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.965 | TFLOPs: 30.21 | 7: iteration 12710/ 21553 | consumed samples: 3253760 | consumed tokens: 6663700480 | elapsed time per iteration (s): 0.30 | learning rate: 8.610E-05 | global batch size: 256 | lm loss: 3.437857E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.153 | TFLOPs: 30.25 | 7: iteration 12720/ 21553 | consumed samples: 3256320 | consumed tokens: 6668943360 | elapsed time per iteration (s): 0.30 | learning rate: 8.597E-05 | global batch size: 256 | lm loss: 3.443223E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.398 | TFLOPs: 30.26 | 7: iteration 12730/ 21553 | consumed samples: 3258880 | consumed tokens: 6674186240 | elapsed time per iteration (s): 0.30 | learning rate: 8.584E-05 | global batch size: 256 | lm loss: 3.433607E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.445 | TFLOPs: 30.26 | 7: iteration 12740/ 21553 | consumed samples: 3261440 | consumed tokens: 6679429120 | elapsed time per iteration (s): 0.30 | learning rate: 8.571E-05 | global batch size: 256 | lm loss: 3.437244E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.243 | TFLOPs: 30.25 | 7: iteration 12750/ 21553 | consumed samples: 3264000 | consumed tokens: 6684672000 | elapsed time per iteration (s): 0.30 | learning rate: 8.559E-05 | global batch size: 256 | lm loss: 3.438937E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.406 | TFLOPs: 30.23 | 7: iteration 12760/ 21553 | consumed samples: 3266560 | consumed tokens: 6689914880 | elapsed time per iteration (s): 0.30 | learning rate: 8.546E-05 | global batch size: 256 | lm loss: 3.438140E+00 | grad norm: 0.460 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.476 | TFLOPs: 30.23 | 7: iteration 12770/ 21553 | consumed samples: 3269120 | consumed tokens: 6695157760 | elapsed time per iteration (s): 0.30 | learning rate: 8.533E-05 | global batch size: 256 | lm loss: 3.439107E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 860.510 | TFLOPs: 30.12 | 7: iteration 12780/ 21553 | consumed samples: 3271680 | consumed tokens: 6700400640 | elapsed time per iteration (s): 0.30 | learning rate: 8.520E-05 | global batch size: 256 | lm loss: 3.435938E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.631 | TFLOPs: 30.23 | 7: iteration 12790/ 21553 | consumed samples: 3274240 | consumed tokens: 6705643520 | elapsed time per iteration (s): 0.30 | learning rate: 8.508E-05 | global batch size: 256 | lm loss: 3.439734E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.409 | TFLOPs: 30.23 | 7: iteration 12800/ 21553 | consumed samples: 3276800 | consumed tokens: 6710886400 | elapsed time per iteration (s): 0.30 | learning rate: 8.495E-05 | global batch size: 256 | lm loss: 3.438552E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.386 | TFLOPs: 30.15 | 7: iteration 12810/ 21553 | consumed samples: 3279360 | consumed tokens: 6716129280 | elapsed time per iteration (s): 0.30 | learning rate: 8.482E-05 | global batch size: 256 | lm loss: 3.442950E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.007 | TFLOPs: 30.25 | 7: iteration 12820/ 21553 | consumed samples: 3281920 | consumed tokens: 6721372160 | elapsed time per iteration (s): 0.30 | learning rate: 8.470E-05 | global batch size: 256 | lm loss: 3.432954E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.952 | TFLOPs: 30.24 | 7: iteration 12830/ 21553 | consumed samples: 3284480 | consumed tokens: 6726615040 | elapsed time per iteration (s): 0.30 | learning rate: 8.457E-05 | global batch size: 256 | lm loss: 3.436275E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.252 | TFLOPs: 30.22 | 7: iteration 12840/ 21553 | consumed samples: 3287040 | consumed tokens: 6731857920 | elapsed time per iteration (s): 0.30 | learning rate: 8.444E-05 | global batch size: 256 | lm loss: 3.429624E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 843.944 | TFLOPs: 29.54 | 7: iteration 12850/ 21553 | consumed samples: 3289600 | consumed tokens: 6737100800 | elapsed time per iteration (s): 0.30 | learning rate: 8.431E-05 | global batch size: 256 | lm loss: 3.433355E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.649 | TFLOPs: 30.27 | 7: iteration 12860/ 21553 | consumed samples: 3292160 | consumed tokens: 6742343680 | elapsed time per iteration (s): 0.30 | learning rate: 8.419E-05 | global batch size: 256 | lm loss: 3.437949E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 846.221 | TFLOPs: 29.62 | 7: iteration 12870/ 21553 | consumed samples: 3294720 | consumed tokens: 6747586560 | elapsed time per iteration (s): 0.30 | learning rate: 8.406E-05 | global batch size: 256 | lm loss: 3.435669E+00 | grad norm: 0.396 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.902 | TFLOPs: 30.21 | 7: iteration 12880/ 21553 | consumed samples: 3297280 | consumed tokens: 6752829440 | elapsed time per iteration (s): 0.30 | learning rate: 8.393E-05 | global batch size: 256 | lm loss: 3.438159E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.649 | TFLOPs: 30.02 | 7: iteration 12890/ 21553 | consumed samples: 3299840 | consumed tokens: 6758072320 | elapsed time per iteration (s): 0.30 | learning rate: 8.381E-05 | global batch size: 256 | lm loss: 3.429840E+00 | grad norm: 0.397 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 848.943 | TFLOPs: 29.72 | 7: iteration 12900/ 21553 | consumed samples: 3302400 | consumed tokens: 6763315200 | elapsed time per iteration (s): 0.30 | learning rate: 8.368E-05 | global batch size: 256 | lm loss: 3.434655E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.320 | TFLOPs: 30.29 | 7: iteration 12910/ 21553 | consumed samples: 3304960 | consumed tokens: 6768558080 | elapsed time per iteration (s): 0.30 | learning rate: 8.355E-05 | global batch size: 256 | lm loss: 3.438809E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.990 | TFLOPs: 30.28 | 7: iteration 12920/ 21553 | consumed samples: 3307520 | consumed tokens: 6773800960 | elapsed time per iteration (s): 0.30 | learning rate: 8.343E-05 | global batch size: 256 | lm loss: 3.437159E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 861.680 | TFLOPs: 30.17 | 7: iteration 12930/ 21553 | consumed samples: 3310080 | consumed tokens: 6779043840 | elapsed time per iteration (s): 0.30 | learning rate: 8.330E-05 | global batch size: 256 | lm loss: 3.439403E+00 | grad norm: 0.465 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.305 | TFLOPs: 30.01 | 7: iteration 12940/ 21553 | consumed samples: 3312640 | consumed tokens: 6784286720 | elapsed time per iteration (s): 0.30 | learning rate: 8.317E-05 | global batch size: 256 | lm loss: 3.435153E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.486 | TFLOPs: 29.98 | 7: iteration 12950/ 21553 | consumed samples: 3315200 | consumed tokens: 6789529600 | elapsed time per iteration (s): 0.37 | learning rate: 8.305E-05 | global batch size: 256 | lm loss: 3.431530E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 684.196 | TFLOPs: 23.95 | 7: iteration 12960/ 21553 | consumed samples: 3317760 | consumed tokens: 6794772480 | elapsed time per iteration (s): 0.36 | learning rate: 8.292E-05 | global batch size: 256 | lm loss: 3.438988E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 715.478 | TFLOPs: 25.05 | 7: iteration 12970/ 21553 | consumed samples: 3320320 | consumed tokens: 6800015360 | elapsed time per iteration (s): 0.33 | learning rate: 8.279E-05 | global batch size: 256 | lm loss: 3.427757E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 768.063 | TFLOPs: 26.89 | 7: iteration 12980/ 21553 | consumed samples: 3322880 | consumed tokens: 6805258240 | elapsed time per iteration (s): 0.35 | learning rate: 8.267E-05 | global batch size: 256 | lm loss: 3.432055E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 734.481 | TFLOPs: 25.71 | 7: iteration 12990/ 21553 | consumed samples: 3325440 | consumed tokens: 6810501120 | elapsed time per iteration (s): 0.32 | learning rate: 8.254E-05 | global batch size: 256 | lm loss: 3.436162E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 803.899 | TFLOPs: 28.14 | 7: iteration 13000/ 21553 | consumed samples: 3328000 | consumed tokens: 6815744000 | elapsed time per iteration (s): 0.30 | learning rate: 8.242E-05 | global batch size: 256 | lm loss: 3.427346E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 859.216 | TFLOPs: 30.08 | 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 13000 | lm loss value: 3.964730E+00 | lm loss PPL: 5.270601E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 13000 to checkpoints_146m14b100mdedup 0: [2023-03-17 01:00:20,405] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step13000 is begin to save! 0: [2023-03-17 01:00:20,410] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:00:20,513] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:00:20,514] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:00:20,530] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:00:20,530] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:00:20,545] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:00:20,546] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:00:20,561] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:00:20,561] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:00:20,577] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:00:20,577] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:00:20,592] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:00:20,593] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:00:20,608] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:00:20,608] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:00:20,624] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:00:20,624] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:00:20,639] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:00:20,640] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:00:20,655] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:00:20,655] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:00:20,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:00:20,671] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:00:20,686] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:00:20,686] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:00:20,701] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:00:20,702] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:00:20,717] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:00:20,717] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:00:20,732] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:00:20,733] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:00:20,748] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:00:20,748] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:00:20,749] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:00:20,750] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step13000/mp_rank_00_model_states.pt 0: [2023-03-17 01:00:20,750] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/mp_rank_00_model_states.pt... 0: [2023-03-17 01:00:20,752] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/mp_rank_00_model_states.pt. 0: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:00:20,770] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:00:20,807] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 3: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:00:20,807] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 3: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:00:20,807] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 2: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:00:20,807] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:00:20,807] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 2: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 2: [2023-03-17 01:00:20,807] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:00:20,808] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 2: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:00:20,808] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 3: [2023-03-17 01:00:20,808] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 2: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 2: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 5: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:00:20,808] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 5: [2023-03-17 01:00:20,808] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 2: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 5: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 5: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:00:20,808] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 1: [2023-03-17 01:00:20,808] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 5: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 1: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 5: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:00:20,808] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 3: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:00:20,808] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:00:20,808] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 0: [2023-03-17 01:00:20,809] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:00:20,809] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 01:00:20,809] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 0: [2023-03-17 01:00:20,809] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:00:20,809] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:00:20,809] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 5: [2023-03-17 01:00:20,809] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 0: [2023-03-17 01:00:20,809] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 5: [2023-03-17 01:00:20,809] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 3: [2023-03-17 01:00:20,809] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:00:20,809] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 01:00:20,809] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 5: [2023-03-17 01:00:20,810] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:00:20,810] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:00:20,810] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 01:00:20,810] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 3: [2023-03-17 01:00:20,810] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:00:20,810] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 01:00:20,810] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 3: [2023-03-17 01:00:20,810] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 01:00:20,810] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 3: [2023-03-17 01:00:20,810] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:00:20,810] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 01:00:20,810] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 3: [2023-03-17 01:00:20,810] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:00:20,811] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 01:00:20,811] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 0: [2023-03-17 01:00:20,811] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:00:20,811] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:00:20,811] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 0: [2023-03-17 01:00:20,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:00:20,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:00:20,813] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 01:00:20,813] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 01:00:20,813] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 0: [2023-03-17 01:00:20,813] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 5: [2023-03-17 01:00:20,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:00:20,813] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 01:00:20,813] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 5: [2023-03-17 01:00:20,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:00:20,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 01:00:20,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 1: [2023-03-17 01:00:20,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:00:20,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 01:00:20,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 1: [2023-03-17 01:00:20,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:00:20,814] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:00:20,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 01:00:20,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 5: [2023-03-17 01:00:20,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 01:00:20,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 2: [2023-03-17 01:00:20,818] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:00:20,819] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:00:20,819] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 2: [2023-03-17 01:00:20,819] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:00:20,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 01:00:20,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:00:20,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:00:20,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 01:00:20,821] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 1: [2023-03-17 01:00:20,821] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 2: [2023-03-17 01:00:20,819] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 01:00:20,819] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:00:20,819] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 2: [2023-03-17 01:00:20,819] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 01:00:20,819] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 0: [2023-03-17 01:00:20,836] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:00:20,836] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:00:20,881] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:00:20,881] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 01:00:20,881] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 01:00:20,881] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:00:20,881] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 01:00:20,881] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 01:00:20,881] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 4: [2023-03-17 01:00:20,881] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 4: [2023-03-17 01:00:20,881] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:00:20,885] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:00:20,885] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 01:00:20,885] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 01:00:20,885] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 01:00:20,885] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 01:00:20,885] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:00:20,885] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 01:00:20,885] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 6: [2023-03-17 01:00:20,885] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:00:20,890] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 01:00:20,890] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 01:00:20,890] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 01:00:20,890] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:00:20,890] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 01:00:20,890] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 01:00:20,890] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:00:20,890] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step13000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 7: [2023-03-17 01:00:20,890] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step13000 is ready now! 0: successfully saved checkpoint at iteration 13000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 538.72 7: iteration 13010/ 21553 | consumed samples: 3330560 | consumed tokens: 6820986880 | elapsed time per iteration (s): 0.36 | learning rate: 8.229E-05 | global batch size: 256 | lm loss: 3.428218E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 707.961 | TFLOPs: 24.78 | 7: iteration 13020/ 21553 | consumed samples: 3333120 | consumed tokens: 6826229760 | elapsed time per iteration (s): 0.31 | learning rate: 8.216E-05 | global batch size: 256 | lm loss: 3.429371E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 835.420 | TFLOPs: 29.25 | 7: iteration 13030/ 21553 | consumed samples: 3335680 | consumed tokens: 6831472640 | elapsed time per iteration (s): 0.30 | learning rate: 8.204E-05 | global batch size: 256 | lm loss: 3.427742E+00 | grad norm: 0.397 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 848.593 | TFLOPs: 29.71 | 7: iteration 13040/ 21553 | consumed samples: 3338240 | consumed tokens: 6836715520 | elapsed time per iteration (s): 0.31 | learning rate: 8.191E-05 | global batch size: 256 | lm loss: 3.436758E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 824.408 | TFLOPs: 28.86 | 7: iteration 13050/ 21553 | consumed samples: 3340800 | consumed tokens: 6841958400 | elapsed time per iteration (s): 0.30 | learning rate: 8.179E-05 | global batch size: 256 | lm loss: 3.424610E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 850.207 | TFLOPs: 29.76 | 7: iteration 13060/ 21553 | consumed samples: 3343360 | consumed tokens: 6847201280 | elapsed time per iteration (s): 0.31 | learning rate: 8.166E-05 | global batch size: 256 | lm loss: 3.436271E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 816.085 | TFLOPs: 28.57 | 7: iteration 13070/ 21553 | consumed samples: 3345920 | consumed tokens: 6852444160 | elapsed time per iteration (s): 0.36 | learning rate: 8.153E-05 | global batch size: 256 | lm loss: 3.428280E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 717.158 | TFLOPs: 25.11 | 7: iteration 13080/ 21553 | consumed samples: 3348480 | consumed tokens: 6857687040 | elapsed time per iteration (s): 0.31 | learning rate: 8.141E-05 | global batch size: 256 | lm loss: 3.428653E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 836.951 | TFLOPs: 29.30 | 7: iteration 13090/ 21553 | consumed samples: 3351040 | consumed tokens: 6862929920 | elapsed time per iteration (s): 0.30 | learning rate: 8.128E-05 | global batch size: 256 | lm loss: 3.434029E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 854.110 | TFLOPs: 29.90 | 7: iteration 13100/ 21553 | consumed samples: 3353600 | consumed tokens: 6868172800 | elapsed time per iteration (s): 0.30 | learning rate: 8.116E-05 | global batch size: 256 | lm loss: 3.419952E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 840.581 | TFLOPs: 29.43 | 7: iteration 13110/ 21553 | consumed samples: 3356160 | consumed tokens: 6873415680 | elapsed time per iteration (s): 0.31 | learning rate: 8.103E-05 | global batch size: 256 | lm loss: 3.432043E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 838.515 | TFLOPs: 29.35 | 7: iteration 13120/ 21553 | consumed samples: 3358720 | consumed tokens: 6878658560 | elapsed time per iteration (s): 0.30 | learning rate: 8.091E-05 | global batch size: 256 | lm loss: 3.434921E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.100 | TFLOPs: 30.00 | 7: iteration 13130/ 21553 | consumed samples: 3361280 | consumed tokens: 6883901440 | elapsed time per iteration (s): 0.31 | learning rate: 8.078E-05 | global batch size: 256 | lm loss: 3.425594E+00 | grad norm: 0.481 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 820.139 | TFLOPs: 28.71 | 7: iteration 13140/ 21553 | consumed samples: 3363840 | consumed tokens: 6889144320 | elapsed time per iteration (s): 0.30 | learning rate: 8.066E-05 | global batch size: 256 | lm loss: 3.427990E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 840.290 | TFLOPs: 29.42 | 7: iteration 13150/ 21553 | consumed samples: 3366400 | consumed tokens: 6894387200 | elapsed time per iteration (s): 0.30 | learning rate: 8.053E-05 | global batch size: 256 | lm loss: 3.428365E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 858.403 | TFLOPs: 30.05 | 7: iteration 13160/ 21553 | consumed samples: 3368960 | consumed tokens: 6899630080 | elapsed time per iteration (s): 0.30 | learning rate: 8.041E-05 | global batch size: 256 | lm loss: 3.423824E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 854.863 | TFLOPs: 29.93 | 7: iteration 13170/ 21553 | consumed samples: 3371520 | consumed tokens: 6904872960 | elapsed time per iteration (s): 0.30 | learning rate: 8.028E-05 | global batch size: 256 | lm loss: 3.431778E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 852.334 | TFLOPs: 29.84 | 7: iteration 13180/ 21553 | consumed samples: 3374080 | consumed tokens: 6910115840 | elapsed time per iteration (s): 0.30 | learning rate: 8.016E-05 | global batch size: 256 | lm loss: 3.427795E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 855.043 | TFLOPs: 29.93 | 7: iteration 13190/ 21553 | consumed samples: 3376640 | consumed tokens: 6915358720 | elapsed time per iteration (s): 0.30 | learning rate: 8.003E-05 | global batch size: 256 | lm loss: 3.433166E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 839.670 | TFLOPs: 29.39 | 7: iteration 13200/ 21553 | consumed samples: 3379200 | consumed tokens: 6920601600 | elapsed time per iteration (s): 0.31 | learning rate: 7.991E-05 | global batch size: 256 | lm loss: 3.426044E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 830.450 | TFLOPs: 29.07 | 7: iteration 13210/ 21553 | consumed samples: 3381760 | consumed tokens: 6925844480 | elapsed time per iteration (s): 0.30 | learning rate: 7.978E-05 | global batch size: 256 | lm loss: 3.424050E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 850.218 | TFLOPs: 29.76 | 7: iteration 13220/ 21553 | consumed samples: 3384320 | consumed tokens: 6931087360 | elapsed time per iteration (s): 0.30 | learning rate: 7.966E-05 | global batch size: 256 | lm loss: 3.427135E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 845.753 | TFLOPs: 29.61 | 7: iteration 13230/ 21553 | consumed samples: 3386880 | consumed tokens: 6936330240 | elapsed time per iteration (s): 0.30 | learning rate: 7.953E-05 | global batch size: 256 | lm loss: 3.424864E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 853.507 | TFLOPs: 29.88 | 7: iteration 13240/ 21553 | consumed samples: 3389440 | consumed tokens: 6941573120 | elapsed time per iteration (s): 0.30 | learning rate: 7.941E-05 | global batch size: 256 | lm loss: 3.421985E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 848.611 | TFLOPs: 29.71 | 7: iteration 13250/ 21553 | consumed samples: 3392000 | consumed tokens: 6946816000 | elapsed time per iteration (s): 0.30 | learning rate: 7.928E-05 | global batch size: 256 | lm loss: 3.428162E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 844.039 | TFLOPs: 29.55 | 7: iteration 13260/ 21553 | consumed samples: 3394560 | consumed tokens: 6952058880 | elapsed time per iteration (s): 0.31 | learning rate: 7.916E-05 | global batch size: 256 | lm loss: 3.428368E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 826.832 | TFLOPs: 28.95 | 7: iteration 13270/ 21553 | consumed samples: 3397120 | consumed tokens: 6957301760 | elapsed time per iteration (s): 0.30 | learning rate: 7.903E-05 | global batch size: 256 | lm loss: 3.425465E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.352 | TFLOPs: 30.19 | 7: iteration 13280/ 21553 | consumed samples: 3399680 | consumed tokens: 6962544640 | elapsed time per iteration (s): 0.30 | learning rate: 7.891E-05 | global batch size: 256 | lm loss: 3.423993E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 843.291 | TFLOPs: 29.52 | 7: iteration 13290/ 21553 | consumed samples: 3402240 | consumed tokens: 6967787520 | elapsed time per iteration (s): 0.30 | learning rate: 7.878E-05 | global batch size: 256 | lm loss: 3.424855E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 839.981 | TFLOPs: 29.41 | 7: iteration 13300/ 21553 | consumed samples: 3404800 | consumed tokens: 6973030400 | elapsed time per iteration (s): 0.31 | learning rate: 7.866E-05 | global batch size: 256 | lm loss: 3.423816E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 816.502 | TFLOPs: 28.58 | 7: iteration 13310/ 21553 | consumed samples: 3407360 | consumed tokens: 6978273280 | elapsed time per iteration (s): 0.31 | learning rate: 7.854E-05 | global batch size: 256 | lm loss: 3.431062E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 835.828 | TFLOPs: 29.26 | 7: iteration 13320/ 21553 | consumed samples: 3409920 | consumed tokens: 6983516160 | elapsed time per iteration (s): 0.30 | learning rate: 7.841E-05 | global batch size: 256 | lm loss: 3.425654E+00 | grad norm: 0.465 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 844.279 | TFLOPs: 29.56 | 7: iteration 13330/ 21553 | consumed samples: 3412480 | consumed tokens: 6988759040 | elapsed time per iteration (s): 0.31 | learning rate: 7.829E-05 | global batch size: 256 | lm loss: 3.425836E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 815.897 | TFLOPs: 28.56 | 7: iteration 13340/ 21553 | consumed samples: 3415040 | consumed tokens: 6994001920 | elapsed time per iteration (s): 0.31 | learning rate: 7.816E-05 | global batch size: 256 | lm loss: 3.423976E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 837.357 | TFLOPs: 29.31 | 7: iteration 13350/ 21553 | consumed samples: 3417600 | consumed tokens: 6999244800 | elapsed time per iteration (s): 0.30 | learning rate: 7.804E-05 | global batch size: 256 | lm loss: 3.425107E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 845.664 | TFLOPs: 29.60 | 7: iteration 13360/ 21553 | consumed samples: 3420160 | consumed tokens: 7004487680 | elapsed time per iteration (s): 0.31 | learning rate: 7.792E-05 | global batch size: 256 | lm loss: 3.427193E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 816.387 | TFLOPs: 28.58 | 7: iteration 13370/ 21553 | consumed samples: 3422720 | consumed tokens: 7009730560 | elapsed time per iteration (s): 0.32 | learning rate: 7.779E-05 | global batch size: 256 | lm loss: 3.423240E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 801.323 | TFLOPs: 28.05 | 7: iteration 13380/ 21553 | consumed samples: 3425280 | consumed tokens: 7014973440 | elapsed time per iteration (s): 0.31 | learning rate: 7.767E-05 | global batch size: 256 | lm loss: 3.416719E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 830.395 | TFLOPs: 29.07 | 7: iteration 13390/ 21553 | consumed samples: 3427840 | consumed tokens: 7020216320 | elapsed time per iteration (s): 0.32 | learning rate: 7.754E-05 | global batch size: 256 | lm loss: 3.424712E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 807.922 | TFLOPs: 28.28 | 7: iteration 13400/ 21553 | consumed samples: 3430400 | consumed tokens: 7025459200 | elapsed time per iteration (s): 0.32 | learning rate: 7.742E-05 | global batch size: 256 | lm loss: 3.425391E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 809.778 | TFLOPs: 28.35 | 7: iteration 13410/ 21553 | consumed samples: 3432960 | consumed tokens: 7030702080 | elapsed time per iteration (s): 0.31 | learning rate: 7.730E-05 | global batch size: 256 | lm loss: 3.428619E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 830.546 | TFLOPs: 29.08 | 7: iteration 13420/ 21553 | consumed samples: 3435520 | consumed tokens: 7035944960 | elapsed time per iteration (s): 0.31 | learning rate: 7.717E-05 | global batch size: 256 | lm loss: 3.427536E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 835.015 | TFLOPs: 29.23 | 7: iteration 13430/ 21553 | consumed samples: 3438080 | consumed tokens: 7041187840 | elapsed time per iteration (s): 0.30 | learning rate: 7.705E-05 | global batch size: 256 | lm loss: 3.423785E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 840.562 | TFLOPs: 29.43 | 7: iteration 13440/ 21553 | consumed samples: 3440640 | consumed tokens: 7046430720 | elapsed time per iteration (s): 0.31 | learning rate: 7.693E-05 | global batch size: 256 | lm loss: 3.422676E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 823.489 | TFLOPs: 28.83 | 7: iteration 13450/ 21553 | consumed samples: 3443200 | consumed tokens: 7051673600 | elapsed time per iteration (s): 0.30 | learning rate: 7.680E-05 | global batch size: 256 | lm loss: 3.417792E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 843.649 | TFLOPs: 29.53 | 7: iteration 13460/ 21553 | consumed samples: 3445760 | consumed tokens: 7056916480 | elapsed time per iteration (s): 0.31 | learning rate: 7.668E-05 | global batch size: 256 | lm loss: 3.422045E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 831.419 | TFLOPs: 29.11 | 7: iteration 13470/ 21553 | consumed samples: 3448320 | consumed tokens: 7062159360 | elapsed time per iteration (s): 0.30 | learning rate: 7.656E-05 | global batch size: 256 | lm loss: 3.418459E+00 | grad norm: 0.458 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 854.390 | TFLOPs: 29.91 | 7: iteration 13480/ 21553 | consumed samples: 3450880 | consumed tokens: 7067402240 | elapsed time per iteration (s): 0.31 | learning rate: 7.644E-05 | global batch size: 256 | lm loss: 3.414005E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 820.192 | TFLOPs: 28.71 | 7: iteration 13490/ 21553 | consumed samples: 3453440 | consumed tokens: 7072645120 | elapsed time per iteration (s): 0.30 | learning rate: 7.631E-05 | global batch size: 256 | lm loss: 3.419074E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 852.519 | TFLOPs: 29.84 | 7: iteration 13500/ 21553 | consumed samples: 3456000 | consumed tokens: 7077888000 | elapsed time per iteration (s): 0.31 | learning rate: 7.619E-05 | global batch size: 256 | lm loss: 3.427307E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 825.923 | TFLOPs: 28.91 | 7: iteration 13510/ 21553 | consumed samples: 3458560 | consumed tokens: 7083130880 | elapsed time per iteration (s): 0.31 | learning rate: 7.607E-05 | global batch size: 256 | lm loss: 3.417209E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 817.866 | TFLOPs: 28.63 | 7: iteration 13520/ 21553 | consumed samples: 3461120 | consumed tokens: 7088373760 | elapsed time per iteration (s): 0.32 | learning rate: 7.594E-05 | global batch size: 256 | lm loss: 3.423087E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 796.629 | TFLOPs: 27.89 | 7: iteration 13530/ 21553 | consumed samples: 3463680 | consumed tokens: 7093616640 | elapsed time per iteration (s): 0.31 | learning rate: 7.582E-05 | global batch size: 256 | lm loss: 3.426725E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 828.726 | TFLOPs: 29.01 | 7: iteration 13540/ 21553 | consumed samples: 3466240 | consumed tokens: 7098859520 | elapsed time per iteration (s): 0.31 | learning rate: 7.570E-05 | global batch size: 256 | lm loss: 3.416420E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 836.041 | TFLOPs: 29.27 | 7: iteration 13550/ 21553 | consumed samples: 3468800 | consumed tokens: 7104102400 | elapsed time per iteration (s): 0.31 | learning rate: 7.558E-05 | global batch size: 256 | lm loss: 3.406777E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 816.314 | TFLOPs: 28.58 | 7: iteration 13560/ 21553 | consumed samples: 3471360 | consumed tokens: 7109345280 | elapsed time per iteration (s): 0.30 | learning rate: 7.545E-05 | global batch size: 256 | lm loss: 3.422794E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 839.553 | TFLOPs: 29.39 | 7: iteration 13570/ 21553 | consumed samples: 3473920 | consumed tokens: 7114588160 | elapsed time per iteration (s): 0.31 | learning rate: 7.533E-05 | global batch size: 256 | lm loss: 3.419621E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 820.821 | TFLOPs: 28.73 | 7: iteration 13580/ 21553 | consumed samples: 3476480 | consumed tokens: 7119831040 | elapsed time per iteration (s): 0.30 | learning rate: 7.521E-05 | global batch size: 256 | lm loss: 3.415356E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.721 | TFLOPs: 30.20 | 7: iteration 13590/ 21553 | consumed samples: 3479040 | consumed tokens: 7125073920 | elapsed time per iteration (s): 0.32 | learning rate: 7.509E-05 | global batch size: 256 | lm loss: 3.422375E+00 | grad norm: 0.451 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 811.853 | TFLOPs: 28.42 | 7: iteration 13600/ 21553 | consumed samples: 3481600 | consumed tokens: 7130316800 | elapsed time per iteration (s): 0.31 | learning rate: 7.497E-05 | global batch size: 256 | lm loss: 3.419221E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 835.095 | TFLOPs: 29.23 | 7: iteration 13610/ 21553 | consumed samples: 3484160 | consumed tokens: 7135559680 | elapsed time per iteration (s): 0.30 | learning rate: 7.484E-05 | global batch size: 256 | lm loss: 3.414046E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.934 | TFLOPs: 30.00 | 7: iteration 13620/ 21553 | consumed samples: 3486720 | consumed tokens: 7140802560 | elapsed time per iteration (s): 0.34 | learning rate: 7.472E-05 | global batch size: 256 | lm loss: 3.420003E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 761.965 | TFLOPs: 26.67 | 7: iteration 13630/ 21553 | consumed samples: 3489280 | consumed tokens: 7146045440 | elapsed time per iteration (s): 0.30 | learning rate: 7.460E-05 | global batch size: 256 | lm loss: 3.418758E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 850.440 | TFLOPs: 29.77 | 7: iteration 13640/ 21553 | consumed samples: 3491840 | consumed tokens: 7151288320 | elapsed time per iteration (s): 0.31 | learning rate: 7.448E-05 | global batch size: 256 | lm loss: 3.417133E+00 | grad norm: 0.449 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 832.779 | TFLOPs: 29.15 | 7: iteration 13650/ 21553 | consumed samples: 3494400 | consumed tokens: 7156531200 | elapsed time per iteration (s): 0.30 | learning rate: 7.436E-05 | global batch size: 256 | lm loss: 3.410810E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 844.116 | TFLOPs: 29.55 | 7: iteration 13660/ 21553 | consumed samples: 3496960 | consumed tokens: 7161774080 | elapsed time per iteration (s): 0.30 | learning rate: 7.423E-05 | global batch size: 256 | lm loss: 3.416586E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.234 | TFLOPs: 29.97 | 7: iteration 13670/ 21553 | consumed samples: 3499520 | consumed tokens: 7167016960 | elapsed time per iteration (s): 0.31 | learning rate: 7.411E-05 | global batch size: 256 | lm loss: 3.416409E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 837.474 | TFLOPs: 29.32 | 7: iteration 13680/ 21553 | consumed samples: 3502080 | consumed tokens: 7172259840 | elapsed time per iteration (s): 0.30 | learning rate: 7.399E-05 | global batch size: 256 | lm loss: 3.412571E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 853.730 | TFLOPs: 29.89 | 7: iteration 13690/ 21553 | consumed samples: 3504640 | consumed tokens: 7177502720 | elapsed time per iteration (s): 0.30 | learning rate: 7.387E-05 | global batch size: 256 | lm loss: 3.418602E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 854.111 | TFLOPs: 29.90 | 7: iteration 13700/ 21553 | consumed samples: 3507200 | consumed tokens: 7182745600 | elapsed time per iteration (s): 0.31 | learning rate: 7.375E-05 | global batch size: 256 | lm loss: 3.416848E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 833.635 | TFLOPs: 29.18 | 7: iteration 13710/ 21553 | consumed samples: 3509760 | consumed tokens: 7187988480 | elapsed time per iteration (s): 0.30 | learning rate: 7.363E-05 | global batch size: 256 | lm loss: 3.419165E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.431 | TFLOPs: 29.98 | 7: iteration 13720/ 21553 | consumed samples: 3512320 | consumed tokens: 7193231360 | elapsed time per iteration (s): 0.30 | learning rate: 7.351E-05 | global batch size: 256 | lm loss: 3.410505E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 843.524 | TFLOPs: 29.53 | 7: iteration 13730/ 21553 | consumed samples: 3514880 | consumed tokens: 7198474240 | elapsed time per iteration (s): 0.31 | learning rate: 7.339E-05 | global batch size: 256 | lm loss: 3.417146E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 836.733 | TFLOPs: 29.29 | 7: iteration 13740/ 21553 | consumed samples: 3517440 | consumed tokens: 7203717120 | elapsed time per iteration (s): 0.30 | learning rate: 7.326E-05 | global batch size: 256 | lm loss: 3.411996E+00 | grad norm: 0.451 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 854.338 | TFLOPs: 29.91 | 7: iteration 13750/ 21553 | consumed samples: 3520000 | consumed tokens: 7208960000 | elapsed time per iteration (s): 0.31 | learning rate: 7.314E-05 | global batch size: 256 | lm loss: 3.414888E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 836.532 | TFLOPs: 29.28 | 7: iteration 13760/ 21553 | consumed samples: 3522560 | consumed tokens: 7214202880 | elapsed time per iteration (s): 0.30 | learning rate: 7.302E-05 | global batch size: 256 | lm loss: 3.412749E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.638 | TFLOPs: 29.99 | 7: iteration 13770/ 21553 | consumed samples: 3525120 | consumed tokens: 7219445760 | elapsed time per iteration (s): 0.30 | learning rate: 7.290E-05 | global batch size: 256 | lm loss: 3.414231E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 847.048 | TFLOPs: 29.65 | 7: iteration 13780/ 21553 | consumed samples: 3527680 | consumed tokens: 7224688640 | elapsed time per iteration (s): 0.30 | learning rate: 7.278E-05 | global batch size: 256 | lm loss: 3.410349E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.851 | TFLOPs: 30.24 | 7: iteration 13790/ 21553 | consumed samples: 3530240 | consumed tokens: 7229931520 | elapsed time per iteration (s): 0.30 | learning rate: 7.266E-05 | global batch size: 256 | lm loss: 3.415982E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.082 | TFLOPs: 30.18 | 7: iteration 13800/ 21553 | consumed samples: 3532800 | consumed tokens: 7235174400 | elapsed time per iteration (s): 0.30 | learning rate: 7.254E-05 | global batch size: 256 | lm loss: 3.412666E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.295 | TFLOPs: 30.26 | 7: iteration 13810/ 21553 | consumed samples: 3535360 | consumed tokens: 7240417280 | elapsed time per iteration (s): 0.30 | learning rate: 7.242E-05 | global batch size: 256 | lm loss: 3.416368E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.409 | TFLOPs: 30.26 | 7: iteration 13820/ 21553 | consumed samples: 3537920 | consumed tokens: 7245660160 | elapsed time per iteration (s): 0.30 | learning rate: 7.230E-05 | global batch size: 256 | lm loss: 3.418986E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.971 | TFLOPs: 30.28 | 7: iteration 13830/ 21553 | consumed samples: 3540480 | consumed tokens: 7250903040 | elapsed time per iteration (s): 0.30 | learning rate: 7.218E-05 | global batch size: 256 | lm loss: 3.421681E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.679 | TFLOPs: 30.27 | 7: iteration 13840/ 21553 | consumed samples: 3543040 | consumed tokens: 7256145920 | elapsed time per iteration (s): 0.30 | learning rate: 7.206E-05 | global batch size: 256 | lm loss: 3.415331E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 848.555 | TFLOPs: 29.71 | 7: iteration 13850/ 21553 | consumed samples: 3545600 | consumed tokens: 7261388800 | elapsed time per iteration (s): 0.30 | learning rate: 7.194E-05 | global batch size: 256 | lm loss: 3.412055E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.460 | TFLOPs: 30.30 | 7: iteration 13860/ 21553 | consumed samples: 3548160 | consumed tokens: 7266631680 | elapsed time per iteration (s): 0.30 | learning rate: 7.182E-05 | global batch size: 256 | lm loss: 3.409264E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.506 | TFLOPs: 30.30 | 7: iteration 13870/ 21553 | consumed samples: 3550720 | consumed tokens: 7271874560 | elapsed time per iteration (s): 0.30 | learning rate: 7.170E-05 | global batch size: 256 | lm loss: 3.406820E+00 | grad norm: 0.454 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.819 | TFLOPs: 30.31 | 7: iteration 13880/ 21553 | consumed samples: 3553280 | consumed tokens: 7277117440 | elapsed time per iteration (s): 0.30 | learning rate: 7.158E-05 | global batch size: 256 | lm loss: 3.413515E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.336 | TFLOPs: 30.29 | 7: iteration 13890/ 21553 | consumed samples: 3555840 | consumed tokens: 7282360320 | elapsed time per iteration (s): 0.30 | learning rate: 7.146E-05 | global batch size: 256 | lm loss: 3.414625E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.007 | TFLOPs: 30.32 | 7: iteration 13900/ 21553 | consumed samples: 3558400 | consumed tokens: 7287603200 | elapsed time per iteration (s): 0.30 | learning rate: 7.134E-05 | global batch size: 256 | lm loss: 3.410813E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.021 | TFLOPs: 29.97 | 7: iteration 13910/ 21553 | consumed samples: 3560960 | consumed tokens: 7292846080 | elapsed time per iteration (s): 0.30 | learning rate: 7.122E-05 | global batch size: 256 | lm loss: 3.413290E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.897 | TFLOPs: 30.28 | 7: iteration 13920/ 21553 | consumed samples: 3563520 | consumed tokens: 7298088960 | elapsed time per iteration (s): 0.30 | learning rate: 7.110E-05 | global batch size: 256 | lm loss: 3.415278E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 842.943 | TFLOPs: 29.51 | 7: iteration 13930/ 21553 | consumed samples: 3566080 | consumed tokens: 7303331840 | elapsed time per iteration (s): 0.30 | learning rate: 7.098E-05 | global batch size: 256 | lm loss: 3.411215E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 853.521 | TFLOPs: 29.88 | 7: iteration 13940/ 21553 | consumed samples: 3568640 | consumed tokens: 7308574720 | elapsed time per iteration (s): 0.30 | learning rate: 7.086E-05 | global batch size: 256 | lm loss: 3.412201E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.323 | TFLOPs: 30.29 | 7: iteration 13950/ 21553 | consumed samples: 3571200 | consumed tokens: 7313817600 | elapsed time per iteration (s): 0.30 | learning rate: 7.074E-05 | global batch size: 256 | lm loss: 3.412566E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.660 | TFLOPs: 30.27 | 7: iteration 13960/ 21553 | consumed samples: 3573760 | consumed tokens: 7319060480 | elapsed time per iteration (s): 0.30 | learning rate: 7.062E-05 | global batch size: 256 | lm loss: 3.415301E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.941 | TFLOPs: 30.28 | 7: iteration 13970/ 21553 | consumed samples: 3576320 | consumed tokens: 7324303360 | elapsed time per iteration (s): 0.30 | learning rate: 7.050E-05 | global batch size: 256 | lm loss: 3.407111E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.596 | TFLOPs: 30.27 | 7: iteration 13980/ 21553 | consumed samples: 3578880 | consumed tokens: 7329546240 | elapsed time per iteration (s): 0.30 | learning rate: 7.038E-05 | global batch size: 256 | lm loss: 3.409381E+00 | grad norm: 0.456 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.279 | TFLOPs: 30.33 | 7: iteration 13990/ 21553 | consumed samples: 3581440 | consumed tokens: 7334789120 | elapsed time per iteration (s): 0.30 | learning rate: 7.027E-05 | global batch size: 256 | lm loss: 3.411911E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.652 | TFLOPs: 30.30 | 0: [2023-03-17 01:05:25,260] [INFO] [logging.py:68:log_dist] [Rank 0] step=14000, skipped=0, lr=[7.014654688611906e-05, 7.014654688611906e-05, 7.014654688611906e-05], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 14000/ 21553 | consumed samples: 3584000 | consumed tokens: 7340032000 | elapsed time per iteration (s): 0.30 | learning rate: 7.015E-05 | global batch size: 256 | lm loss: 3.408196E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.460 | TFLOPs: 30.19 | 0: steps: 14000 loss: 3.3912 iter time (s): 0.300 samples/sec: 853.719 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 14000 | lm loss value: 3.922349E+00 | lm loss PPL: 5.051899E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 14000 to checkpoints_146m14b100mdedup 0: [2023-03-17 01:05:25,404] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step14000 is begin to save! 0: [2023-03-17 01:05:25,409] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:05:25,503] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:05:25,504] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:05:25,520] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:05:25,520] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:05:25,535] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:05:25,535] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:05:25,550] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:05:25,551] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:05:25,565] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:05:25,566] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:05:25,580] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:05:25,581] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:05:25,595] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:05:25,596] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:05:25,610] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:05:25,610] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:05:25,625] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:05:25,625] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:05:25,640] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:05:25,640] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:05:25,655] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:05:25,655] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:05:25,670] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:05:25,671] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:05:25,685] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:05:25,686] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:05:25,700] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:05:25,701] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:05:25,715] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:05:25,716] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:05:25,730] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:05:25,730] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:05:25,732] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:05:25,732] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step14000/mp_rank_00_model_states.pt 0: [2023-03-17 01:05:25,732] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/mp_rank_00_model_states.pt... 0: [2023-03-17 01:05:25,734] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/mp_rank_00_model_states.pt. 0: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:05:25,752] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:05:25,784] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:05:25,784] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,784] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 0: [2023-03-17 01:05:25,784] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,785] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,785] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:05:25,785] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 0: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:05:25,786] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 4: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:05:25,786] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,786] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 4: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 6: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,786] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 4: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:05:25,786] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 7: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:05:25,786] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:05:25,786] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 01:05:25,787] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 7: [2023-03-17 01:05:25,787] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 01:05:25,787] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 0: [2023-03-17 01:05:25,787] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:05:25,787] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,787] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 4: [2023-03-17 01:05:25,787] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:05:25,787] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 01:05:25,788] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 0: [2023-03-17 01:05:25,789] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:05:25,789] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,789] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 6: [2023-03-17 01:05:25,789] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,789] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,789] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 01:05:25,789] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:05:25,789] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 6: [2023-03-17 01:05:25,789] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 6: [2023-03-17 01:05:25,790] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,790] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 01:05:25,790] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 4: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:05:25,791] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 6: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,791] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,791] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 6: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 6: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,791] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 0: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:05:25,792] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,792] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 7: [2023-03-17 01:05:25,789] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:05:25,789] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 01:05:25,789] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 7: [2023-03-17 01:05:25,789] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:05:25,790] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 01:05:25,790] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 7: [2023-03-17 01:05:25,790] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:05:25,790] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:05:25,790] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 7: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:05:25,791] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 01:05:25,791] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 7: [2023-03-17 01:05:25,792] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:05:25,792] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 01:05:25,792] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 7: [2023-03-17 01:05:25,792] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:05:25,792] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:05:25,792] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 4: [2023-03-17 01:05:25,794] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:05:25,794] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:05:25,795] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 4: [2023-03-17 01:05:25,795] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:05:25,795] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 01:05:25,795] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 4: [2023-03-17 01:05:25,797] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:05:25,797] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:05:25,797] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 01:05:25,797] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 01:05:25,797] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 4: [2023-03-17 01:05:25,797] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 2: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:05:25,800] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 2: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:05:25,800] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 2: [2023-03-17 01:05:25,800] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 01:05:25,800] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 01:05:25,800] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 2: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 2: [2023-03-17 01:05:25,800] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 2: [2023-03-17 01:05:25,801] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:05:25,801] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:05:25,801] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:05:25,801] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:05:25,801] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 01:05:25,801] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:05:25,801] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 01:05:25,801] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:05:25,801] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 2: [2023-03-17 01:05:25,801] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 2: [2023-03-17 01:05:25,801] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 2: [2023-03-17 01:05:25,801] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:05:25,804] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:05:25,804] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:05:25,804] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 1: [2023-03-17 01:05:25,804] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 1: [2023-03-17 01:05:25,804] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 1: [2023-03-17 01:05:25,804] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 01:05:25,804] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 01:05:25,804] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 1: [2023-03-17 01:05:25,804] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 5: [2023-03-17 01:05:25,805] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:05:25,805] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 01:05:25,805] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 5: [2023-03-17 01:05:25,805] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:05:25,805] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 01:05:25,806] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 5: [2023-03-17 01:05:25,806] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:05:25,807] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 01:05:25,807] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 5: [2023-03-17 01:05:25,807] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:05:25,807] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 01:05:25,807] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 5: [2023-03-17 01:05:25,809] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:05:25,809] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:05:25,809] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:05:25,810] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 01:05:25,810] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 01:05:25,809] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:05:25,810] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 5: [2023-03-17 01:05:25,810] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 5: [2023-03-17 01:05:25,810] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 01:05:25,810] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 01:05:25,810] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 5: [2023-03-17 01:05:25,810] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 0: [2023-03-17 01:05:25,814] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:05:25,814] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:05:25,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 01:05:25,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 01:05:25,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 01:05:25,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 01:05:25,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 01:05:25,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:05:25,815] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 3: [2023-03-17 01:05:25,815] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 3: [2023-03-17 01:05:25,816] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:05:25,817] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step14000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 01:05:25,817] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step14000 is ready now! 0: successfully saved checkpoint at iteration 14000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 440.35 7: iteration 14010/ 21553 | consumed samples: 3586560 | consumed tokens: 7345274880 | elapsed time per iteration (s): 0.35 | learning rate: 7.003E-05 | global batch size: 256 | lm loss: 3.411329E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 729.005 | TFLOPs: 25.52 | 7: iteration 14020/ 21553 | consumed samples: 3589120 | consumed tokens: 7350517760 | elapsed time per iteration (s): 0.30 | learning rate: 6.991E-05 | global batch size: 256 | lm loss: 3.411670E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.233 | TFLOPs: 30.32 | 7: iteration 14030/ 21553 | consumed samples: 3591680 | consumed tokens: 7355760640 | elapsed time per iteration (s): 0.30 | learning rate: 6.979E-05 | global batch size: 256 | lm loss: 3.407659E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.262 | TFLOPs: 30.33 | 7: iteration 14040/ 21553 | consumed samples: 3594240 | consumed tokens: 7361003520 | elapsed time per iteration (s): 0.30 | learning rate: 6.967E-05 | global batch size: 256 | lm loss: 3.409484E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.277 | TFLOPs: 30.33 | 7: iteration 14050/ 21553 | consumed samples: 3596800 | consumed tokens: 7366246400 | elapsed time per iteration (s): 0.30 | learning rate: 6.955E-05 | global batch size: 256 | lm loss: 3.412632E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.851 | TFLOPs: 30.35 | 7: iteration 14060/ 21553 | consumed samples: 3599360 | consumed tokens: 7371489280 | elapsed time per iteration (s): 0.30 | learning rate: 6.944E-05 | global batch size: 256 | lm loss: 3.409879E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.559 | TFLOPs: 30.34 | 7: iteration 14070/ 21553 | consumed samples: 3601920 | consumed tokens: 7376732160 | elapsed time per iteration (s): 0.30 | learning rate: 6.932E-05 | global batch size: 256 | lm loss: 3.411879E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.526 | TFLOPs: 30.33 | 7: iteration 14080/ 21553 | consumed samples: 3604480 | consumed tokens: 7381975040 | elapsed time per iteration (s): 0.30 | learning rate: 6.920E-05 | global batch size: 256 | lm loss: 3.411698E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.906 | TFLOPs: 30.31 | 7: iteration 14090/ 21553 | consumed samples: 3607040 | consumed tokens: 7387217920 | elapsed time per iteration (s): 0.30 | learning rate: 6.908E-05 | global batch size: 256 | lm loss: 3.401847E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.269 | TFLOPs: 30.33 | 7: iteration 14100/ 21553 | consumed samples: 3609600 | consumed tokens: 7392460800 | elapsed time per iteration (s): 0.30 | learning rate: 6.896E-05 | global batch size: 256 | lm loss: 3.400793E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.915 | TFLOPs: 30.31 | 7: iteration 14110/ 21553 | consumed samples: 3612160 | consumed tokens: 7397703680 | elapsed time per iteration (s): 0.30 | learning rate: 6.884E-05 | global batch size: 256 | lm loss: 3.407980E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 859.946 | TFLOPs: 30.10 | 7: iteration 14120/ 21553 | consumed samples: 3614720 | consumed tokens: 7402946560 | elapsed time per iteration (s): 0.30 | learning rate: 6.873E-05 | global batch size: 256 | lm loss: 3.401411E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.826 | TFLOPs: 30.31 | 7: iteration 14130/ 21553 | consumed samples: 3617280 | consumed tokens: 7408189440 | elapsed time per iteration (s): 0.30 | learning rate: 6.861E-05 | global batch size: 256 | lm loss: 3.406124E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 855.919 | TFLOPs: 29.96 | 7: iteration 14140/ 21553 | consumed samples: 3619840 | consumed tokens: 7413432320 | elapsed time per iteration (s): 0.31 | learning rate: 6.849E-05 | global batch size: 256 | lm loss: 3.411339E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 834.959 | TFLOPs: 29.23 | 7: iteration 14150/ 21553 | consumed samples: 3622400 | consumed tokens: 7418675200 | elapsed time per iteration (s): 0.30 | learning rate: 6.837E-05 | global batch size: 256 | lm loss: 3.403886E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.791 | TFLOPs: 30.31 | 7: iteration 14160/ 21553 | consumed samples: 3624960 | consumed tokens: 7423918080 | elapsed time per iteration (s): 0.30 | learning rate: 6.826E-05 | global batch size: 256 | lm loss: 3.409494E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.644 | TFLOPs: 30.30 | 7: iteration 14170/ 21553 | consumed samples: 3627520 | consumed tokens: 7429160960 | elapsed time per iteration (s): 0.30 | learning rate: 6.814E-05 | global batch size: 256 | lm loss: 3.408953E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.283 | TFLOPs: 30.22 | 7: iteration 14180/ 21553 | consumed samples: 3630080 | consumed tokens: 7434403840 | elapsed time per iteration (s): 0.30 | learning rate: 6.802E-05 | global batch size: 256 | lm loss: 3.410711E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.439 | TFLOPs: 30.33 | 7: iteration 14190/ 21553 | consumed samples: 3632640 | consumed tokens: 7439646720 | elapsed time per iteration (s): 0.30 | learning rate: 6.791E-05 | global batch size: 256 | lm loss: 3.404402E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.998 | TFLOPs: 30.32 | 7: iteration 14200/ 21553 | consumed samples: 3635200 | consumed tokens: 7444889600 | elapsed time per iteration (s): 0.30 | learning rate: 6.779E-05 | global batch size: 256 | lm loss: 3.406046E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.184 | TFLOPs: 30.32 | 7: iteration 14210/ 21553 | consumed samples: 3637760 | consumed tokens: 7450132480 | elapsed time per iteration (s): 0.30 | learning rate: 6.767E-05 | global batch size: 256 | lm loss: 3.402760E+00 | grad norm: 0.459 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.973 | TFLOPs: 30.32 | 7: iteration 14220/ 21553 | consumed samples: 3640320 | consumed tokens: 7455375360 | elapsed time per iteration (s): 0.30 | learning rate: 6.755E-05 | global batch size: 256 | lm loss: 3.409461E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.169 | TFLOPs: 30.29 | 7: iteration 14230/ 21553 | consumed samples: 3642880 | consumed tokens: 7460618240 | elapsed time per iteration (s): 0.30 | learning rate: 6.744E-05 | global batch size: 256 | lm loss: 3.406499E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.748 | TFLOPs: 29.99 | 7: iteration 14240/ 21553 | consumed samples: 3645440 | consumed tokens: 7465861120 | elapsed time per iteration (s): 0.30 | learning rate: 6.732E-05 | global batch size: 256 | lm loss: 3.408647E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.238 | TFLOPs: 30.01 | 7: iteration 14250/ 21553 | consumed samples: 3648000 | consumed tokens: 7471104000 | elapsed time per iteration (s): 0.30 | learning rate: 6.720E-05 | global batch size: 256 | lm loss: 3.407371E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.916 | TFLOPs: 30.31 | 7: iteration 14260/ 21553 | consumed samples: 3650560 | consumed tokens: 7476346880 | elapsed time per iteration (s): 0.30 | learning rate: 6.709E-05 | global batch size: 256 | lm loss: 3.400788E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.306 | TFLOPs: 30.33 | 7: iteration 14270/ 21553 | consumed samples: 3653120 | consumed tokens: 7481589760 | elapsed time per iteration (s): 0.30 | learning rate: 6.697E-05 | global batch size: 256 | lm loss: 3.404721E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.318 | TFLOPs: 30.29 | 7: iteration 14280/ 21553 | consumed samples: 3655680 | consumed tokens: 7486832640 | elapsed time per iteration (s): 0.30 | learning rate: 6.685E-05 | global batch size: 256 | lm loss: 3.402669E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.481 | TFLOPs: 30.30 | 7: iteration 14290/ 21553 | consumed samples: 3658240 | consumed tokens: 7492075520 | elapsed time per iteration (s): 0.30 | learning rate: 6.674E-05 | global batch size: 256 | lm loss: 3.401260E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.069 | TFLOPs: 30.28 | 7: iteration 14300/ 21553 | consumed samples: 3660800 | consumed tokens: 7497318400 | elapsed time per iteration (s): 0.30 | learning rate: 6.662E-05 | global batch size: 256 | lm loss: 3.408952E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.138 | TFLOPs: 30.32 | 7: iteration 14310/ 21553 | consumed samples: 3663360 | consumed tokens: 7502561280 | elapsed time per iteration (s): 0.30 | learning rate: 6.651E-05 | global batch size: 256 | lm loss: 3.408649E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.762 | TFLOPs: 30.31 | 7: iteration 14320/ 21553 | consumed samples: 3665920 | consumed tokens: 7507804160 | elapsed time per iteration (s): 0.30 | learning rate: 6.639E-05 | global batch size: 256 | lm loss: 3.400711E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.947 | TFLOPs: 30.31 | 7: iteration 14330/ 21553 | consumed samples: 3668480 | consumed tokens: 7513047040 | elapsed time per iteration (s): 0.30 | learning rate: 6.627E-05 | global batch size: 256 | lm loss: 3.409076E+00 | grad norm: 0.455 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.997 | TFLOPs: 30.32 | 7: iteration 14340/ 21553 | consumed samples: 3671040 | consumed tokens: 7518289920 | elapsed time per iteration (s): 0.30 | learning rate: 6.616E-05 | global batch size: 256 | lm loss: 3.402862E+00 | grad norm: 0.449 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.268 | TFLOPs: 30.33 | 7: iteration 14350/ 21553 | consumed samples: 3673600 | consumed tokens: 7523532800 | elapsed time per iteration (s): 0.30 | learning rate: 6.604E-05 | global batch size: 256 | lm loss: 3.403401E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.673 | TFLOPs: 30.30 | 7: iteration 14360/ 21553 | consumed samples: 3676160 | consumed tokens: 7528775680 | elapsed time per iteration (s): 0.30 | learning rate: 6.593E-05 | global batch size: 256 | lm loss: 3.398638E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.775 | TFLOPs: 30.27 | 7: iteration 14370/ 21553 | consumed samples: 3678720 | consumed tokens: 7534018560 | elapsed time per iteration (s): 0.30 | learning rate: 6.581E-05 | global batch size: 256 | lm loss: 3.405235E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.810 | TFLOPs: 30.31 | 7: iteration 14380/ 21553 | consumed samples: 3681280 | consumed tokens: 7539261440 | elapsed time per iteration (s): 0.30 | learning rate: 6.570E-05 | global batch size: 256 | lm loss: 3.398200E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.246 | TFLOPs: 30.29 | 7: iteration 14390/ 21553 | consumed samples: 3683840 | consumed tokens: 7544504320 | elapsed time per iteration (s): 0.30 | learning rate: 6.558E-05 | global batch size: 256 | lm loss: 3.397940E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.620 | TFLOPs: 30.30 | 7: iteration 14400/ 21553 | consumed samples: 3686400 | consumed tokens: 7549747200 | elapsed time per iteration (s): 0.30 | learning rate: 6.547E-05 | global batch size: 256 | lm loss: 3.400603E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.559 | TFLOPs: 30.30 | 7: iteration 14410/ 21553 | consumed samples: 3688960 | consumed tokens: 7554990080 | elapsed time per iteration (s): 0.30 | learning rate: 6.535E-05 | global batch size: 256 | lm loss: 3.400898E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.982 | TFLOPs: 30.32 | 7: iteration 14420/ 21553 | consumed samples: 3691520 | consumed tokens: 7560232960 | elapsed time per iteration (s): 0.30 | learning rate: 6.524E-05 | global batch size: 256 | lm loss: 3.398661E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.521 | TFLOPs: 30.30 | 7: iteration 14430/ 21553 | consumed samples: 3694080 | consumed tokens: 7565475840 | elapsed time per iteration (s): 0.30 | learning rate: 6.512E-05 | global batch size: 256 | lm loss: 3.401994E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.368 | TFLOPs: 30.29 | 7: iteration 14440/ 21553 | consumed samples: 3696640 | consumed tokens: 7570718720 | elapsed time per iteration (s): 0.30 | learning rate: 6.501E-05 | global batch size: 256 | lm loss: 3.405311E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.798 | TFLOPs: 30.31 | 7: iteration 14450/ 21553 | consumed samples: 3699200 | consumed tokens: 7575961600 | elapsed time per iteration (s): 0.30 | learning rate: 6.489E-05 | global batch size: 256 | lm loss: 3.396418E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.157 | TFLOPs: 30.32 | 7: iteration 14460/ 21553 | consumed samples: 3701760 | consumed tokens: 7581204480 | elapsed time per iteration (s): 0.30 | learning rate: 6.478E-05 | global batch size: 256 | lm loss: 3.402298E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.575 | TFLOPs: 30.30 | 7: iteration 14470/ 21553 | consumed samples: 3704320 | consumed tokens: 7586447360 | elapsed time per iteration (s): 0.30 | learning rate: 6.466E-05 | global batch size: 256 | lm loss: 3.401720E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.518 | TFLOPs: 30.33 | 7: iteration 14480/ 21553 | consumed samples: 3706880 | consumed tokens: 7591690240 | elapsed time per iteration (s): 0.30 | learning rate: 6.455E-05 | global batch size: 256 | lm loss: 3.393081E+00 | grad norm: 0.401 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.930 | TFLOPs: 30.35 | 7: iteration 14490/ 21553 | consumed samples: 3709440 | consumed tokens: 7596933120 | elapsed time per iteration (s): 0.30 | learning rate: 6.443E-05 | global batch size: 256 | lm loss: 3.403349E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.769 | TFLOPs: 30.34 | 7: iteration 14500/ 21553 | consumed samples: 3712000 | consumed tokens: 7602176000 | elapsed time per iteration (s): 0.30 | learning rate: 6.432E-05 | global batch size: 256 | lm loss: 3.404420E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.994 | TFLOPs: 30.32 | 7: iteration 14510/ 21553 | consumed samples: 3714560 | consumed tokens: 7607418880 | elapsed time per iteration (s): 0.30 | learning rate: 6.421E-05 | global batch size: 256 | lm loss: 3.394195E+00 | grad norm: 0.398 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.817 | TFLOPs: 30.31 | 7: iteration 14520/ 21553 | consumed samples: 3717120 | consumed tokens: 7612661760 | elapsed time per iteration (s): 0.30 | learning rate: 6.409E-05 | global batch size: 256 | lm loss: 3.397697E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.918 | TFLOPs: 30.03 | 7: iteration 14530/ 21553 | consumed samples: 3719680 | consumed tokens: 7617904640 | elapsed time per iteration (s): 0.30 | learning rate: 6.398E-05 | global batch size: 256 | lm loss: 3.397915E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.904 | TFLOPs: 30.31 | 7: iteration 14540/ 21553 | consumed samples: 3722240 | consumed tokens: 7623147520 | elapsed time per iteration (s): 0.30 | learning rate: 6.386E-05 | global batch size: 256 | lm loss: 3.400824E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.437 | TFLOPs: 30.30 | 7: iteration 14550/ 21553 | consumed samples: 3724800 | consumed tokens: 7628390400 | elapsed time per iteration (s): 0.30 | learning rate: 6.375E-05 | global batch size: 256 | lm loss: 3.397068E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.398 | TFLOPs: 30.30 | 7: iteration 14560/ 21553 | consumed samples: 3727360 | consumed tokens: 7633633280 | elapsed time per iteration (s): 0.30 | learning rate: 6.364E-05 | global batch size: 256 | lm loss: 3.398138E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.627 | TFLOPs: 30.30 | 7: iteration 14570/ 21553 | consumed samples: 3729920 | consumed tokens: 7638876160 | elapsed time per iteration (s): 0.30 | learning rate: 6.352E-05 | global batch size: 256 | lm loss: 3.397878E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.486 | TFLOPs: 30.33 | 7: iteration 14580/ 21553 | consumed samples: 3732480 | consumed tokens: 7644119040 | elapsed time per iteration (s): 0.30 | learning rate: 6.341E-05 | global batch size: 256 | lm loss: 3.396543E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.894 | TFLOPs: 30.31 | 7: iteration 14590/ 21553 | consumed samples: 3735040 | consumed tokens: 7649361920 | elapsed time per iteration (s): 0.30 | learning rate: 6.330E-05 | global batch size: 256 | lm loss: 3.394619E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.685 | TFLOPs: 30.34 | 7: iteration 14600/ 21553 | consumed samples: 3737600 | consumed tokens: 7654604800 | elapsed time per iteration (s): 0.30 | learning rate: 6.318E-05 | global batch size: 256 | lm loss: 3.395198E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.361 | TFLOPs: 30.33 | 7: iteration 14610/ 21553 | consumed samples: 3740160 | consumed tokens: 7659847680 | elapsed time per iteration (s): 0.30 | learning rate: 6.307E-05 | global batch size: 256 | lm loss: 3.401442E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.446 | TFLOPs: 30.33 | 7: iteration 14620/ 21553 | consumed samples: 3742720 | consumed tokens: 7665090560 | elapsed time per iteration (s): 0.30 | learning rate: 6.296E-05 | global batch size: 256 | lm loss: 3.398479E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.143 | TFLOPs: 30.32 | 7: iteration 14630/ 21553 | consumed samples: 3745280 | consumed tokens: 7670333440 | elapsed time per iteration (s): 0.30 | learning rate: 6.284E-05 | global batch size: 256 | lm loss: 3.399449E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.080 | TFLOPs: 30.32 | 7: iteration 14640/ 21553 | consumed samples: 3747840 | consumed tokens: 7675576320 | elapsed time per iteration (s): 0.30 | learning rate: 6.273E-05 | global batch size: 256 | lm loss: 3.402422E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.452 | TFLOPs: 30.33 | 7: iteration 14650/ 21553 | consumed samples: 3750400 | consumed tokens: 7680819200 | elapsed time per iteration (s): 0.30 | learning rate: 6.262E-05 | global batch size: 256 | lm loss: 3.400337E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.906 | TFLOPs: 30.31 | 7: iteration 14660/ 21553 | consumed samples: 3752960 | consumed tokens: 7686062080 | elapsed time per iteration (s): 0.30 | learning rate: 6.251E-05 | global batch size: 256 | lm loss: 3.394287E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.062 | TFLOPs: 30.32 | 7: iteration 14670/ 21553 | consumed samples: 3755520 | consumed tokens: 7691304960 | elapsed time per iteration (s): 0.30 | learning rate: 6.239E-05 | global batch size: 256 | lm loss: 3.396855E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.365 | TFLOPs: 30.33 | 7: iteration 14680/ 21553 | consumed samples: 3758080 | consumed tokens: 7696547840 | elapsed time per iteration (s): 0.30 | learning rate: 6.228E-05 | global batch size: 256 | lm loss: 3.391538E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.320 | TFLOPs: 30.33 | 7: iteration 14690/ 21553 | consumed samples: 3760640 | consumed tokens: 7701790720 | elapsed time per iteration (s): 0.30 | learning rate: 6.217E-05 | global batch size: 256 | lm loss: 3.395970E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 857.495 | TFLOPs: 30.02 | 7: iteration 14700/ 21553 | consumed samples: 3763200 | consumed tokens: 7707033600 | elapsed time per iteration (s): 0.30 | learning rate: 6.206E-05 | global batch size: 256 | lm loss: 3.388347E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.271 | TFLOPs: 30.33 | 7: iteration 14710/ 21553 | consumed samples: 3765760 | consumed tokens: 7712276480 | elapsed time per iteration (s): 0.30 | learning rate: 6.194E-05 | global batch size: 256 | lm loss: 3.394488E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.282 | TFLOPs: 30.33 | 7: iteration 14720/ 21553 | consumed samples: 3768320 | consumed tokens: 7717519360 | elapsed time per iteration (s): 0.30 | learning rate: 6.183E-05 | global batch size: 256 | lm loss: 3.398017E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.870 | TFLOPs: 30.31 | 7: iteration 14730/ 21553 | consumed samples: 3770880 | consumed tokens: 7722762240 | elapsed time per iteration (s): 0.30 | learning rate: 6.172E-05 | global batch size: 256 | lm loss: 3.400655E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.377 | TFLOPs: 30.33 | 7: iteration 14740/ 21553 | consumed samples: 3773440 | consumed tokens: 7728005120 | elapsed time per iteration (s): 0.30 | learning rate: 6.161E-05 | global batch size: 256 | lm loss: 3.396972E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.418 | TFLOPs: 30.33 | 7: iteration 14750/ 21553 | consumed samples: 3776000 | consumed tokens: 7733248000 | elapsed time per iteration (s): 0.30 | learning rate: 6.150E-05 | global batch size: 256 | lm loss: 3.398213E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.431 | TFLOPs: 30.33 | 7: iteration 14760/ 21553 | consumed samples: 3778560 | consumed tokens: 7738490880 | elapsed time per iteration (s): 0.30 | learning rate: 6.139E-05 | global batch size: 256 | lm loss: 3.390897E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.345 | TFLOPs: 30.33 | 7: iteration 14770/ 21553 | consumed samples: 3781120 | consumed tokens: 7743733760 | elapsed time per iteration (s): 0.30 | learning rate: 6.127E-05 | global batch size: 256 | lm loss: 3.390076E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.201 | TFLOPs: 30.32 | 7: iteration 14780/ 21553 | consumed samples: 3783680 | consumed tokens: 7748976640 | elapsed time per iteration (s): 0.30 | learning rate: 6.116E-05 | global batch size: 256 | lm loss: 3.395661E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.795 | TFLOPs: 30.31 | 7: iteration 14790/ 21553 | consumed samples: 3786240 | consumed tokens: 7754219520 | elapsed time per iteration (s): 0.30 | learning rate: 6.105E-05 | global batch size: 256 | lm loss: 3.392900E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.929 | TFLOPs: 30.31 | 7: iteration 14800/ 21553 | consumed samples: 3788800 | consumed tokens: 7759462400 | elapsed time per iteration (s): 0.30 | learning rate: 6.094E-05 | global batch size: 256 | lm loss: 3.398233E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.960 | TFLOPs: 30.31 | 7: iteration 14810/ 21553 | consumed samples: 3791360 | consumed tokens: 7764705280 | elapsed time per iteration (s): 0.30 | learning rate: 6.083E-05 | global batch size: 256 | lm loss: 3.390508E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.115 | TFLOPs: 30.32 | 7: iteration 14820/ 21553 | consumed samples: 3793920 | consumed tokens: 7769948160 | elapsed time per iteration (s): 0.30 | learning rate: 6.072E-05 | global batch size: 256 | lm loss: 3.403014E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.163 | TFLOPs: 30.32 | 7: iteration 14830/ 21553 | consumed samples: 3796480 | consumed tokens: 7775191040 | elapsed time per iteration (s): 0.30 | learning rate: 6.061E-05 | global batch size: 256 | lm loss: 3.395078E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.396 | TFLOPs: 30.30 | 7: iteration 14840/ 21553 | consumed samples: 3799040 | consumed tokens: 7780433920 | elapsed time per iteration (s): 0.30 | learning rate: 6.050E-05 | global batch size: 256 | lm loss: 3.389360E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.946 | TFLOPs: 30.31 | 7: iteration 14850/ 21553 | consumed samples: 3801600 | consumed tokens: 7785676800 | elapsed time per iteration (s): 0.30 | learning rate: 6.039E-05 | global batch size: 256 | lm loss: 3.397297E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.925 | TFLOPs: 30.31 | 7: iteration 14860/ 21553 | consumed samples: 3804160 | consumed tokens: 7790919680 | elapsed time per iteration (s): 0.30 | learning rate: 6.028E-05 | global batch size: 256 | lm loss: 3.384859E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.305 | TFLOPs: 30.33 | 7: iteration 14870/ 21553 | consumed samples: 3806720 | consumed tokens: 7796162560 | elapsed time per iteration (s): 0.30 | learning rate: 6.016E-05 | global batch size: 256 | lm loss: 3.389577E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.142 | TFLOPs: 30.32 | 7: iteration 14880/ 21553 | consumed samples: 3809280 | consumed tokens: 7801405440 | elapsed time per iteration (s): 0.30 | learning rate: 6.005E-05 | global batch size: 256 | lm loss: 3.396927E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.064 | TFLOPs: 30.32 | 7: iteration 14890/ 21553 | consumed samples: 3811840 | consumed tokens: 7806648320 | elapsed time per iteration (s): 0.30 | learning rate: 5.994E-05 | global batch size: 256 | lm loss: 3.393304E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.709 | TFLOPs: 30.34 | 7: iteration 14900/ 21553 | consumed samples: 3814400 | consumed tokens: 7811891200 | elapsed time per iteration (s): 0.30 | learning rate: 5.983E-05 | global batch size: 256 | lm loss: 3.391718E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.265 | TFLOPs: 30.33 | 7: iteration 14910/ 21553 | consumed samples: 3816960 | consumed tokens: 7817134080 | elapsed time per iteration (s): 0.30 | learning rate: 5.972E-05 | global batch size: 256 | lm loss: 3.386605E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.442 | TFLOPs: 30.33 | 7: iteration 14920/ 21553 | consumed samples: 3819520 | consumed tokens: 7822376960 | elapsed time per iteration (s): 0.30 | learning rate: 5.961E-05 | global batch size: 256 | lm loss: 3.388197E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.807 | TFLOPs: 30.34 | 7: iteration 14930/ 21553 | consumed samples: 3822080 | consumed tokens: 7827619840 | elapsed time per iteration (s): 0.30 | learning rate: 5.950E-05 | global batch size: 256 | lm loss: 3.385294E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.750 | TFLOPs: 30.34 | 7: iteration 14940/ 21553 | consumed samples: 3824640 | consumed tokens: 7832862720 | elapsed time per iteration (s): 0.30 | learning rate: 5.940E-05 | global batch size: 256 | lm loss: 3.393939E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.285 | TFLOPs: 30.33 | 7: iteration 14950/ 21553 | consumed samples: 3827200 | consumed tokens: 7838105600 | elapsed time per iteration (s): 0.30 | learning rate: 5.929E-05 | global batch size: 256 | lm loss: 3.396705E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.858 | TFLOPs: 30.35 | 7: iteration 14960/ 21553 | consumed samples: 3829760 | consumed tokens: 7843348480 | elapsed time per iteration (s): 0.30 | learning rate: 5.918E-05 | global batch size: 256 | lm loss: 3.393931E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.571 | TFLOPs: 30.34 | 7: iteration 14970/ 21553 | consumed samples: 3832320 | consumed tokens: 7848591360 | elapsed time per iteration (s): 0.30 | learning rate: 5.907E-05 | global batch size: 256 | lm loss: 3.392849E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.971 | TFLOPs: 30.35 | 7: iteration 14980/ 21553 | consumed samples: 3834880 | consumed tokens: 7853834240 | elapsed time per iteration (s): 0.30 | learning rate: 5.896E-05 | global batch size: 256 | lm loss: 3.394931E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.353 | TFLOPs: 30.33 | 7: iteration 14990/ 21553 | consumed samples: 3837440 | consumed tokens: 7859077120 | elapsed time per iteration (s): 0.30 | learning rate: 5.885E-05 | global batch size: 256 | lm loss: 3.390854E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.168 | TFLOPs: 30.32 | 7: iteration 15000/ 21553 | consumed samples: 3840000 | consumed tokens: 7864320000 | elapsed time per iteration (s): 0.30 | learning rate: 5.874E-05 | global batch size: 256 | lm loss: 3.391375E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.376 | TFLOPs: 30.33 | 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 15000 | lm loss value: 3.960290E+00 | lm loss PPL: 5.247254E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 15000 to checkpoints_146m14b100mdedup 0: [2023-03-17 01:10:21,803] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step15000 is begin to save! 0: [2023-03-17 01:10:21,807] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:10:21,901] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:10:21,902] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:10:21,917] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:10:21,917] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:10:21,932] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:10:21,932] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:10:21,947] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:10:21,948] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:10:21,962] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:10:21,963] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:10:21,977] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:10:21,978] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:10:21,993] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:10:21,993] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:10:22,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:10:22,008] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:10:22,023] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:10:22,023] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:10:22,038] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:10:22,038] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:10:22,053] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:10:22,053] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:10:22,068] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:10:22,069] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:10:22,083] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:10:22,083] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:10:22,098] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:10:22,098] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:10:22,113] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:10:22,114] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:10:22,128] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:10:22,128] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:10:22,129] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:10:22,130] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step15000/mp_rank_00_model_states.pt 0: [2023-03-17 01:10:22,130] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/mp_rank_00_model_states.pt... 0: [2023-03-17 01:10:22,134] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/mp_rank_00_model_states.pt. 0: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:10:22,151] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:10:22,184] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:10:22,184] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:10:22,184] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:10:22,184] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 0: [2023-03-17 01:10:22,186] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:10:22,187] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 01:10:22,187] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 0: [2023-03-17 01:10:22,189] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:10:22,189] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 01:10:22,189] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 4: [2023-03-17 01:10:22,189] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:10:22,189] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:10:22,189] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 01:10:22,189] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 01:10:22,189] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 4: [2023-03-17 01:10:22,189] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 0: [2023-03-17 01:10:22,191] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:10:22,191] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 4: [2023-03-17 01:10:22,191] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:10:22,191] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 4: [2023-03-17 01:10:22,191] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 01:10:22,191] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 0: [2023-03-17 01:10:22,192] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:10:22,192] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 01:10:22,192] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 4: [2023-03-17 01:10:22,192] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:10:22,192] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 0: [2023-03-17 01:10:22,192] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:10:22,192] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 0: [2023-03-17 01:10:22,192] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 01:10:22,192] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 6: [2023-03-17 01:10:22,192] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:10:22,193] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:10:22,193] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 01:10:22,194] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 4: [2023-03-17 01:10:22,194] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:10:22,194] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:10:22,194] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 7: [2023-03-17 01:10:22,194] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:10:22,194] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 01:10:22,194] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 7: [2023-03-17 01:10:22,194] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:10:22,194] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:10:22,194] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:10:22,195] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:10:22,195] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 01:10:22,195] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 01:10:22,195] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 7: [2023-03-17 01:10:22,195] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 7: [2023-03-17 01:10:22,195] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 0: [2023-03-17 01:10:22,196] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:10:22,196] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 01:10:22,196] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 7: [2023-03-17 01:10:22,197] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:10:22,197] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:10:22,198] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:10:22,198] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 7: [2023-03-17 01:10:22,198] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 01:10:22,198] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 7: [2023-03-17 01:10:22,198] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:10:22,198] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 01:10:22,198] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 7: [2023-03-17 01:10:22,198] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:10:22,198] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 01:10:22,198] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 4: [2023-03-17 01:10:22,201] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:10:22,201] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:10:22,201] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 01:10:22,201] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 01:10:22,202] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 4: [2023-03-17 01:10:22,202] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 6: [2023-03-17 01:10:22,193] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 01:10:22,193] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 6: [2023-03-17 01:10:22,193] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:10:22,193] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:10:22,193] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 6: [2023-03-17 01:10:22,193] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:10:22,193] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:10:22,193] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 6: [2023-03-17 01:10:22,195] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:10:22,195] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 01:10:22,195] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 6: [2023-03-17 01:10:22,195] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:10:22,195] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 01:10:22,195] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 6: [2023-03-17 01:10:22,195] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:10:22,196] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 01:10:22,196] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 6: [2023-03-17 01:10:22,199] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:10:22,199] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 01:10:22,199] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:10:22,199] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 6: [2023-03-17 01:10:22,199] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 01:10:22,199] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:10:22,206] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:10:22,206] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 01:10:22,206] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 01:10:22,206] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:10:22,206] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:10:22,206] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 2: [2023-03-17 01:10:22,206] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 01:10:22,206] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 2: [2023-03-17 01:10:22,206] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:10:22,209] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 01:10:22,209] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 01:10:22,209] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 01:10:22,209] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 3: [2023-03-17 01:10:22,209] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:10:22,209] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 01:10:22,209] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 01:10:22,209] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 3: [2023-03-17 01:10:22,209] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 1: [2023-03-17 01:10:22,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:10:22,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:10:22,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:10:22,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:10:22,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:10:22,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:10:22,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:10:22,210] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:10:22,211] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 01:10:22,211] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 01:10:22,211] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:10:22,211] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 1: [2023-03-17 01:10:22,211] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 01:10:22,211] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 01:10:22,211] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 01:10:22,211] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:10:22,211] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 1: [2023-03-17 01:10:22,211] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 01:10:22,211] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 1: [2023-03-17 01:10:22,211] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 1: [2023-03-17 01:10:22,211] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 1: [2023-03-17 01:10:22,211] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 1: [2023-03-17 01:10:22,211] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 1: [2023-03-17 01:10:22,211] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 0: [2023-03-17 01:10:22,213] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:10:22,213] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:10:22,221] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 01:10:22,221] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 01:10:22,221] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 01:10:22,221] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 01:10:22,221] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 01:10:22,221] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 01:10:22,221] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 01:10:22,221] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step15000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 5: [2023-03-17 01:10:22,221] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step15000 is ready now! 0: successfully saved checkpoint at iteration 15000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 424.73 7: iteration 15010/ 21553 | consumed samples: 3842560 | consumed tokens: 7869562880 | elapsed time per iteration (s): 0.35 | learning rate: 5.863E-05 | global batch size: 256 | lm loss: 3.391102E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 736.971 | TFLOPs: 25.80 | 7: iteration 15020/ 21553 | consumed samples: 3845120 | consumed tokens: 7874805760 | elapsed time per iteration (s): 0.30 | learning rate: 5.852E-05 | global batch size: 256 | lm loss: 3.389394E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.281 | TFLOPs: 30.36 | 7: iteration 15030/ 21553 | consumed samples: 3847680 | consumed tokens: 7880048640 | elapsed time per iteration (s): 0.30 | learning rate: 5.841E-05 | global batch size: 256 | lm loss: 3.395507E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.885 | TFLOPs: 30.35 | 7: iteration 15040/ 21553 | consumed samples: 3850240 | consumed tokens: 7885291520 | elapsed time per iteration (s): 0.30 | learning rate: 5.830E-05 | global batch size: 256 | lm loss: 3.385669E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.057 | TFLOPs: 30.35 | 7: iteration 15050/ 21553 | consumed samples: 3852800 | consumed tokens: 7890534400 | elapsed time per iteration (s): 0.30 | learning rate: 5.820E-05 | global batch size: 256 | lm loss: 3.390759E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.549 | TFLOPs: 30.34 | 7: iteration 15060/ 21553 | consumed samples: 3855360 | consumed tokens: 7895777280 | elapsed time per iteration (s): 0.30 | learning rate: 5.809E-05 | global batch size: 256 | lm loss: 3.395061E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.602 | TFLOPs: 30.34 | 7: iteration 15070/ 21553 | consumed samples: 3857920 | consumed tokens: 7901020160 | elapsed time per iteration (s): 0.30 | learning rate: 5.798E-05 | global batch size: 256 | lm loss: 3.381477E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.725 | TFLOPs: 30.34 | 7: iteration 15080/ 21553 | consumed samples: 3860480 | consumed tokens: 7906263040 | elapsed time per iteration (s): 0.30 | learning rate: 5.787E-05 | global batch size: 256 | lm loss: 3.393375E+00 | grad norm: 0.462 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.188 | TFLOPs: 30.36 | 7: iteration 15090/ 21553 | consumed samples: 3863040 | consumed tokens: 7911505920 | elapsed time per iteration (s): 0.30 | learning rate: 5.776E-05 | global batch size: 256 | lm loss: 3.393617E+00 | grad norm: 0.467 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.172 | TFLOPs: 30.36 | 7: iteration 15100/ 21553 | consumed samples: 3865600 | consumed tokens: 7916748800 | elapsed time per iteration (s): 0.30 | learning rate: 5.766E-05 | global batch size: 256 | lm loss: 3.377112E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.863 | TFLOPs: 30.35 | 7: iteration 15110/ 21553 | consumed samples: 3868160 | consumed tokens: 7921991680 | elapsed time per iteration (s): 0.30 | learning rate: 5.755E-05 | global batch size: 256 | lm loss: 3.391363E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.141 | TFLOPs: 30.32 | 7: iteration 15120/ 21553 | consumed samples: 3870720 | consumed tokens: 7927234560 | elapsed time per iteration (s): 0.30 | learning rate: 5.744E-05 | global batch size: 256 | lm loss: 3.389478E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.109 | TFLOPs: 30.32 | 7: iteration 15130/ 21553 | consumed samples: 3873280 | consumed tokens: 7932477440 | elapsed time per iteration (s): 0.30 | learning rate: 5.733E-05 | global batch size: 256 | lm loss: 3.388450E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.921 | TFLOPs: 30.31 | 7: iteration 15140/ 21553 | consumed samples: 3875840 | consumed tokens: 7937720320 | elapsed time per iteration (s): 0.30 | learning rate: 5.723E-05 | global batch size: 256 | lm loss: 3.391102E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.097 | TFLOPs: 30.32 | 7: iteration 15150/ 21553 | consumed samples: 3878400 | consumed tokens: 7942963200 | elapsed time per iteration (s): 0.30 | learning rate: 5.712E-05 | global batch size: 256 | lm loss: 3.383557E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.727 | TFLOPs: 30.34 | 7: iteration 15160/ 21553 | consumed samples: 3880960 | consumed tokens: 7948206080 | elapsed time per iteration (s): 0.30 | learning rate: 5.701E-05 | global batch size: 256 | lm loss: 3.387844E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.107 | TFLOPs: 30.36 | 7: iteration 15170/ 21553 | consumed samples: 3883520 | consumed tokens: 7953448960 | elapsed time per iteration (s): 0.30 | learning rate: 5.690E-05 | global batch size: 256 | lm loss: 3.390662E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.215 | TFLOPs: 30.36 | 7: iteration 15180/ 21553 | consumed samples: 3886080 | consumed tokens: 7958691840 | elapsed time per iteration (s): 0.30 | learning rate: 5.680E-05 | global batch size: 256 | lm loss: 3.386794E+00 | grad norm: 0.447 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.661 | TFLOPs: 30.37 | 7: iteration 15190/ 21553 | consumed samples: 3888640 | consumed tokens: 7963934720 | elapsed time per iteration (s): 0.30 | learning rate: 5.669E-05 | global batch size: 256 | lm loss: 3.390020E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.101 | TFLOPs: 30.35 | 7: iteration 15200/ 21553 | consumed samples: 3891200 | consumed tokens: 7969177600 | elapsed time per iteration (s): 0.30 | learning rate: 5.658E-05 | global batch size: 256 | lm loss: 3.387154E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.900 | TFLOPs: 30.35 | 7: iteration 15210/ 21553 | consumed samples: 3893760 | consumed tokens: 7974420480 | elapsed time per iteration (s): 0.30 | learning rate: 5.648E-05 | global batch size: 256 | lm loss: 3.386830E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.821 | TFLOPs: 30.34 | 7: iteration 15220/ 21553 | consumed samples: 3896320 | consumed tokens: 7979663360 | elapsed time per iteration (s): 0.30 | learning rate: 5.637E-05 | global batch size: 256 | lm loss: 3.382756E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.830 | TFLOPs: 30.35 | 7: iteration 15230/ 21553 | consumed samples: 3898880 | consumed tokens: 7984906240 | elapsed time per iteration (s): 0.30 | learning rate: 5.626E-05 | global batch size: 256 | lm loss: 3.391493E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.709 | TFLOPs: 30.34 | 7: iteration 15240/ 21553 | consumed samples: 3901440 | consumed tokens: 7990149120 | elapsed time per iteration (s): 0.30 | learning rate: 5.616E-05 | global batch size: 256 | lm loss: 3.390260E+00 | grad norm: 0.454 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.874 | TFLOPs: 30.31 | 7: iteration 15250/ 21553 | consumed samples: 3904000 | consumed tokens: 7995392000 | elapsed time per iteration (s): 0.30 | learning rate: 5.605E-05 | global batch size: 256 | lm loss: 3.386906E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.140 | TFLOPs: 30.32 | 7: iteration 15260/ 21553 | consumed samples: 3906560 | consumed tokens: 8000634880 | elapsed time per iteration (s): 0.30 | learning rate: 5.595E-05 | global batch size: 256 | lm loss: 3.390169E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.110 | TFLOPs: 30.32 | 7: iteration 15270/ 21553 | consumed samples: 3909120 | consumed tokens: 8005877760 | elapsed time per iteration (s): 0.30 | learning rate: 5.584E-05 | global batch size: 256 | lm loss: 3.386877E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.845 | TFLOPs: 30.31 | 7: iteration 15280/ 21553 | consumed samples: 3911680 | consumed tokens: 8011120640 | elapsed time per iteration (s): 0.30 | learning rate: 5.573E-05 | global batch size: 256 | lm loss: 3.387091E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.770 | TFLOPs: 30.31 | 7: iteration 15290/ 21553 | consumed samples: 3914240 | consumed tokens: 8016363520 | elapsed time per iteration (s): 0.30 | learning rate: 5.563E-05 | global batch size: 256 | lm loss: 3.386422E+00 | grad norm: 0.454 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.855 | TFLOPs: 30.31 | 7: iteration 15300/ 21553 | consumed samples: 3916800 | consumed tokens: 8021606400 | elapsed time per iteration (s): 0.30 | learning rate: 5.552E-05 | global batch size: 256 | lm loss: 3.386302E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.313 | TFLOPs: 30.33 | 7: iteration 15310/ 21553 | consumed samples: 3919360 | consumed tokens: 8026849280 | elapsed time per iteration (s): 0.30 | learning rate: 5.542E-05 | global batch size: 256 | lm loss: 3.386714E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.174 | TFLOPs: 30.32 | 7: iteration 15320/ 21553 | consumed samples: 3921920 | consumed tokens: 8032092160 | elapsed time per iteration (s): 0.30 | learning rate: 5.531E-05 | global batch size: 256 | lm loss: 3.382830E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.928 | TFLOPs: 30.31 | 7: iteration 15330/ 21553 | consumed samples: 3924480 | consumed tokens: 8037335040 | elapsed time per iteration (s): 0.30 | learning rate: 5.521E-05 | global batch size: 256 | lm loss: 3.381070E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.177 | TFLOPs: 30.32 | 7: iteration 15340/ 21553 | consumed samples: 3927040 | consumed tokens: 8042577920 | elapsed time per iteration (s): 0.30 | learning rate: 5.510E-05 | global batch size: 256 | lm loss: 3.389404E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.865 | TFLOPs: 30.31 | 7: iteration 15350/ 21553 | consumed samples: 3929600 | consumed tokens: 8047820800 | elapsed time per iteration (s): 0.30 | learning rate: 5.500E-05 | global batch size: 256 | lm loss: 3.383082E+00 | grad norm: 0.449 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.153 | TFLOPs: 30.32 | 7: iteration 15360/ 21553 | consumed samples: 3932160 | consumed tokens: 8053063680 | elapsed time per iteration (s): 0.30 | learning rate: 5.489E-05 | global batch size: 256 | lm loss: 3.384172E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.204 | TFLOPs: 30.32 | 7: iteration 15370/ 21553 | consumed samples: 3934720 | consumed tokens: 8058306560 | elapsed time per iteration (s): 0.30 | learning rate: 5.479E-05 | global batch size: 256 | lm loss: 3.379964E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.987 | TFLOPs: 30.32 | 7: iteration 15380/ 21553 | consumed samples: 3937280 | consumed tokens: 8063549440 | elapsed time per iteration (s): 0.30 | learning rate: 5.468E-05 | global batch size: 256 | lm loss: 3.385035E+00 | grad norm: 0.449 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.048 | TFLOPs: 30.32 | 7: iteration 15390/ 21553 | consumed samples: 3939840 | consumed tokens: 8068792320 | elapsed time per iteration (s): 0.30 | learning rate: 5.458E-05 | global batch size: 256 | lm loss: 3.382492E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.950 | TFLOPs: 30.31 | 7: iteration 15400/ 21553 | consumed samples: 3942400 | consumed tokens: 8074035200 | elapsed time per iteration (s): 0.30 | learning rate: 5.447E-05 | global batch size: 256 | lm loss: 3.387411E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.995 | TFLOPs: 30.32 | 7: iteration 15410/ 21553 | consumed samples: 3944960 | consumed tokens: 8079278080 | elapsed time per iteration (s): 0.30 | learning rate: 5.437E-05 | global batch size: 256 | lm loss: 3.382707E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.462 | TFLOPs: 30.30 | 7: iteration 15420/ 21553 | consumed samples: 3947520 | consumed tokens: 8084520960 | elapsed time per iteration (s): 0.30 | learning rate: 5.427E-05 | global batch size: 256 | lm loss: 3.383319E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.017 | TFLOPs: 30.32 | 7: iteration 15430/ 21553 | consumed samples: 3950080 | consumed tokens: 8089763840 | elapsed time per iteration (s): 0.30 | learning rate: 5.416E-05 | global batch size: 256 | lm loss: 3.379562E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.750 | TFLOPs: 30.31 | 7: iteration 15440/ 21553 | consumed samples: 3952640 | consumed tokens: 8095006720 | elapsed time per iteration (s): 0.30 | learning rate: 5.406E-05 | global batch size: 256 | lm loss: 3.382052E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.207 | TFLOPs: 30.29 | 7: iteration 15450/ 21553 | consumed samples: 3955200 | consumed tokens: 8100249600 | elapsed time per iteration (s): 0.30 | learning rate: 5.395E-05 | global batch size: 256 | lm loss: 3.380721E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.592 | TFLOPs: 30.30 | 7: iteration 15460/ 21553 | consumed samples: 3957760 | consumed tokens: 8105492480 | elapsed time per iteration (s): 0.30 | learning rate: 5.385E-05 | global batch size: 256 | lm loss: 3.377237E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.862 | TFLOPs: 30.31 | 7: iteration 15470/ 21553 | consumed samples: 3960320 | consumed tokens: 8110735360 | elapsed time per iteration (s): 0.30 | learning rate: 5.375E-05 | global batch size: 256 | lm loss: 3.385667E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.877 | TFLOPs: 30.31 | 7: iteration 15480/ 21553 | consumed samples: 3962880 | consumed tokens: 8115978240 | elapsed time per iteration (s): 0.30 | learning rate: 5.364E-05 | global batch size: 256 | lm loss: 3.379555E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.691 | TFLOPs: 30.31 | 7: iteration 15490/ 21553 | consumed samples: 3965440 | consumed tokens: 8121221120 | elapsed time per iteration (s): 0.30 | learning rate: 5.354E-05 | global batch size: 256 | lm loss: 3.380797E+00 | grad norm: 0.456 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.939 | TFLOPs: 30.31 | 7: iteration 15500/ 21553 | consumed samples: 3968000 | consumed tokens: 8126464000 | elapsed time per iteration (s): 0.30 | learning rate: 5.344E-05 | global batch size: 256 | lm loss: 3.384625E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.770 | TFLOPs: 30.31 | 7: iteration 15510/ 21553 | consumed samples: 3970560 | consumed tokens: 8131706880 | elapsed time per iteration (s): 0.30 | learning rate: 5.333E-05 | global batch size: 256 | lm loss: 3.385575E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.100 | TFLOPs: 30.32 | 7: iteration 15520/ 21553 | consumed samples: 3973120 | consumed tokens: 8136949760 | elapsed time per iteration (s): 0.30 | learning rate: 5.323E-05 | global batch size: 256 | lm loss: 3.386400E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.664 | TFLOPs: 30.30 | 7: iteration 15530/ 21553 | consumed samples: 3975680 | consumed tokens: 8142192640 | elapsed time per iteration (s): 0.30 | learning rate: 5.313E-05 | global batch size: 256 | lm loss: 3.380094E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.698 | TFLOPs: 30.31 | 7: iteration 15540/ 21553 | consumed samples: 3978240 | consumed tokens: 8147435520 | elapsed time per iteration (s): 0.30 | learning rate: 5.303E-05 | global batch size: 256 | lm loss: 3.376199E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.658 | TFLOPs: 30.30 | 7: iteration 15550/ 21553 | consumed samples: 3980800 | consumed tokens: 8152678400 | elapsed time per iteration (s): 0.30 | learning rate: 5.292E-05 | global batch size: 256 | lm loss: 3.383435E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.046 | TFLOPs: 30.32 | 7: iteration 15560/ 21553 | consumed samples: 3983360 | consumed tokens: 8157921280 | elapsed time per iteration (s): 0.30 | learning rate: 5.282E-05 | global batch size: 256 | lm loss: 3.379220E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.826 | TFLOPs: 30.31 | 7: iteration 15570/ 21553 | consumed samples: 3985920 | consumed tokens: 8163164160 | elapsed time per iteration (s): 0.30 | learning rate: 5.272E-05 | global batch size: 256 | lm loss: 3.375607E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.913 | TFLOPs: 30.31 | 7: iteration 15580/ 21553 | consumed samples: 3988480 | consumed tokens: 8168407040 | elapsed time per iteration (s): 0.30 | learning rate: 5.262E-05 | global batch size: 256 | lm loss: 3.376091E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.933 | TFLOPs: 30.31 | 7: iteration 15590/ 21553 | consumed samples: 3991040 | consumed tokens: 8173649920 | elapsed time per iteration (s): 0.30 | learning rate: 5.251E-05 | global batch size: 256 | lm loss: 3.376503E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.432 | TFLOPs: 30.30 | 7: iteration 15600/ 21553 | consumed samples: 3993600 | consumed tokens: 8178892800 | elapsed time per iteration (s): 0.30 | learning rate: 5.241E-05 | global batch size: 256 | lm loss: 3.387214E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.041 | TFLOPs: 30.28 | 7: iteration 15610/ 21553 | consumed samples: 3996160 | consumed tokens: 8184135680 | elapsed time per iteration (s): 0.30 | learning rate: 5.231E-05 | global batch size: 256 | lm loss: 3.378371E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.704 | TFLOPs: 30.31 | 7: iteration 15620/ 21553 | consumed samples: 3998720 | consumed tokens: 8189378560 | elapsed time per iteration (s): 0.30 | learning rate: 5.221E-05 | global batch size: 256 | lm loss: 3.375901E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.171 | TFLOPs: 30.32 | 7: iteration 15630/ 21553 | consumed samples: 4001280 | consumed tokens: 8194621440 | elapsed time per iteration (s): 0.30 | learning rate: 5.211E-05 | global batch size: 256 | lm loss: 3.374197E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.909 | TFLOPs: 30.31 | 7: iteration 15640/ 21553 | consumed samples: 4003840 | consumed tokens: 8199864320 | elapsed time per iteration (s): 0.30 | learning rate: 5.201E-05 | global batch size: 256 | lm loss: 3.382780E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.763 | TFLOPs: 30.31 | 7: iteration 15650/ 21553 | consumed samples: 4006400 | consumed tokens: 8205107200 | elapsed time per iteration (s): 0.30 | learning rate: 5.191E-05 | global batch size: 256 | lm loss: 3.384566E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.485 | TFLOPs: 30.30 | 7: iteration 15660/ 21553 | consumed samples: 4008960 | consumed tokens: 8210350080 | elapsed time per iteration (s): 0.30 | learning rate: 5.180E-05 | global batch size: 256 | lm loss: 3.377985E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.462 | TFLOPs: 30.30 | 7: iteration 15670/ 21553 | consumed samples: 4011520 | consumed tokens: 8215592960 | elapsed time per iteration (s): 0.30 | learning rate: 5.170E-05 | global batch size: 256 | lm loss: 3.380850E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.620 | TFLOPs: 30.30 | 7: iteration 15680/ 21553 | consumed samples: 4014080 | consumed tokens: 8220835840 | elapsed time per iteration (s): 0.30 | learning rate: 5.160E-05 | global batch size: 256 | lm loss: 3.373172E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.437 | TFLOPs: 30.30 | 7: iteration 15690/ 21553 | consumed samples: 4016640 | consumed tokens: 8226078720 | elapsed time per iteration (s): 0.30 | learning rate: 5.150E-05 | global batch size: 256 | lm loss: 3.381007E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.910 | TFLOPs: 30.31 | 7: iteration 15700/ 21553 | consumed samples: 4019200 | consumed tokens: 8231321600 | elapsed time per iteration (s): 0.30 | learning rate: 5.140E-05 | global batch size: 256 | lm loss: 3.378830E+00 | grad norm: 0.469 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.122 | TFLOPs: 30.32 | 7: iteration 15710/ 21553 | consumed samples: 4021760 | consumed tokens: 8236564480 | elapsed time per iteration (s): 0.30 | learning rate: 5.130E-05 | global batch size: 256 | lm loss: 3.374897E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.911 | TFLOPs: 30.31 | 7: iteration 15720/ 21553 | consumed samples: 4024320 | consumed tokens: 8241807360 | elapsed time per iteration (s): 0.30 | learning rate: 5.120E-05 | global batch size: 256 | lm loss: 3.377894E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.663 | TFLOPs: 30.30 | 7: iteration 15730/ 21553 | consumed samples: 4026880 | consumed tokens: 8247050240 | elapsed time per iteration (s): 0.30 | learning rate: 5.110E-05 | global batch size: 256 | lm loss: 3.370824E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.912 | TFLOPs: 30.31 | 7: iteration 15740/ 21553 | consumed samples: 4029440 | consumed tokens: 8252293120 | elapsed time per iteration (s): 0.30 | learning rate: 5.100E-05 | global batch size: 256 | lm loss: 3.378494E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.587 | TFLOPs: 30.34 | 7: iteration 15750/ 21553 | consumed samples: 4032000 | consumed tokens: 8257536000 | elapsed time per iteration (s): 0.30 | learning rate: 5.090E-05 | global batch size: 256 | lm loss: 3.373381E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.373 | TFLOPs: 30.33 | 7: iteration 15760/ 21553 | consumed samples: 4034560 | consumed tokens: 8262778880 | elapsed time per iteration (s): 0.30 | learning rate: 5.080E-05 | global batch size: 256 | lm loss: 3.385530E+00 | grad norm: 0.455 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.959 | TFLOPs: 30.28 | 7: iteration 15770/ 21553 | consumed samples: 4037120 | consumed tokens: 8268021760 | elapsed time per iteration (s): 0.30 | learning rate: 5.070E-05 | global batch size: 256 | lm loss: 3.380088E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.500 | TFLOPs: 30.33 | 7: iteration 15780/ 21553 | consumed samples: 4039680 | consumed tokens: 8273264640 | elapsed time per iteration (s): 0.30 | learning rate: 5.060E-05 | global batch size: 256 | lm loss: 3.378144E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.303 | TFLOPs: 30.33 | 7: iteration 15790/ 21553 | consumed samples: 4042240 | consumed tokens: 8278507520 | elapsed time per iteration (s): 0.30 | learning rate: 5.050E-05 | global batch size: 256 | lm loss: 3.378686E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.941 | TFLOPs: 30.31 | 7: iteration 15800/ 21553 | consumed samples: 4044800 | consumed tokens: 8283750400 | elapsed time per iteration (s): 0.30 | learning rate: 5.040E-05 | global batch size: 256 | lm loss: 3.378231E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.060 | TFLOPs: 30.32 | 7: iteration 15810/ 21553 | consumed samples: 4047360 | consumed tokens: 8288993280 | elapsed time per iteration (s): 0.30 | learning rate: 5.030E-05 | global batch size: 256 | lm loss: 3.377513E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.360 | TFLOPs: 30.33 | 7: iteration 15820/ 21553 | consumed samples: 4049920 | consumed tokens: 8294236160 | elapsed time per iteration (s): 0.30 | learning rate: 5.020E-05 | global batch size: 256 | lm loss: 3.370828E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.447 | TFLOPs: 30.33 | 7: iteration 15830/ 21553 | consumed samples: 4052480 | consumed tokens: 8299479040 | elapsed time per iteration (s): 0.30 | learning rate: 5.010E-05 | global batch size: 256 | lm loss: 3.370324E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.250 | TFLOPs: 30.33 | 7: iteration 15840/ 21553 | consumed samples: 4055040 | consumed tokens: 8304721920 | elapsed time per iteration (s): 0.30 | learning rate: 5.001E-05 | global batch size: 256 | lm loss: 3.374316E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.077 | TFLOPs: 30.32 | 7: iteration 15850/ 21553 | consumed samples: 4057600 | consumed tokens: 8309964800 | elapsed time per iteration (s): 0.30 | learning rate: 4.991E-05 | global batch size: 256 | lm loss: 3.380489E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.190 | TFLOPs: 30.32 | 7: iteration 15860/ 21553 | consumed samples: 4060160 | consumed tokens: 8315207680 | elapsed time per iteration (s): 0.30 | learning rate: 4.981E-05 | global batch size: 256 | lm loss: 3.378601E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.439 | TFLOPs: 30.30 | 7: iteration 15870/ 21553 | consumed samples: 4062720 | consumed tokens: 8320450560 | elapsed time per iteration (s): 0.30 | learning rate: 4.971E-05 | global batch size: 256 | lm loss: 3.368218E+00 | grad norm: 0.448 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.827 | TFLOPs: 30.31 | 7: iteration 15880/ 21553 | consumed samples: 4065280 | consumed tokens: 8325693440 | elapsed time per iteration (s): 0.30 | learning rate: 4.961E-05 | global batch size: 256 | lm loss: 3.374764E+00 | grad norm: 0.461 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.787 | TFLOPs: 30.31 | 7: iteration 15890/ 21553 | consumed samples: 4067840 | consumed tokens: 8330936320 | elapsed time per iteration (s): 0.30 | learning rate: 4.951E-05 | global batch size: 256 | lm loss: 3.373978E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.161 | TFLOPs: 30.32 | 7: iteration 15900/ 21553 | consumed samples: 4070400 | consumed tokens: 8336179200 | elapsed time per iteration (s): 0.30 | learning rate: 4.942E-05 | global batch size: 256 | lm loss: 3.374540E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.195 | TFLOPs: 30.32 | 7: iteration 15910/ 21553 | consumed samples: 4072960 | consumed tokens: 8341422080 | elapsed time per iteration (s): 0.30 | learning rate: 4.932E-05 | global batch size: 256 | lm loss: 3.375720E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.799 | TFLOPs: 30.31 | 7: iteration 15920/ 21553 | consumed samples: 4075520 | consumed tokens: 8346664960 | elapsed time per iteration (s): 0.30 | learning rate: 4.922E-05 | global batch size: 256 | lm loss: 3.373762E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.985 | TFLOPs: 30.32 | 7: iteration 15930/ 21553 | consumed samples: 4078080 | consumed tokens: 8351907840 | elapsed time per iteration (s): 0.30 | learning rate: 4.912E-05 | global batch size: 256 | lm loss: 3.366096E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.169 | TFLOPs: 30.32 | 7: iteration 15940/ 21553 | consumed samples: 4080640 | consumed tokens: 8357150720 | elapsed time per iteration (s): 0.30 | learning rate: 4.902E-05 | global batch size: 256 | lm loss: 3.373995E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.402 | TFLOPs: 30.33 | 7: iteration 15950/ 21553 | consumed samples: 4083200 | consumed tokens: 8362393600 | elapsed time per iteration (s): 0.30 | learning rate: 4.893E-05 | global batch size: 256 | lm loss: 3.373800E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.004 | TFLOPs: 30.32 | 7: iteration 15960/ 21553 | consumed samples: 4085760 | consumed tokens: 8367636480 | elapsed time per iteration (s): 0.30 | learning rate: 4.883E-05 | global batch size: 256 | lm loss: 3.366102E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.504 | TFLOPs: 30.33 | 7: iteration 15970/ 21553 | consumed samples: 4088320 | consumed tokens: 8372879360 | elapsed time per iteration (s): 0.30 | learning rate: 4.873E-05 | global batch size: 256 | lm loss: 3.379571E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.461 | TFLOPs: 30.33 | 7: iteration 15980/ 21553 | consumed samples: 4090880 | consumed tokens: 8378122240 | elapsed time per iteration (s): 0.30 | learning rate: 4.864E-05 | global batch size: 256 | lm loss: 3.375568E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.229 | TFLOPs: 30.32 | 7: iteration 15990/ 21553 | consumed samples: 4093440 | consumed tokens: 8383365120 | elapsed time per iteration (s): 0.30 | learning rate: 4.854E-05 | global batch size: 256 | lm loss: 3.376061E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.137 | TFLOPs: 30.32 | 0: [2023-03-17 01:15:17,787] [INFO] [logging.py:68:log_dist] [Rank 0] step=16000, skipped=0, lr=[4.8441849544340955e-05, 4.8441849544340955e-05, 4.8441849544340955e-05], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 16000/ 21553 | consumed samples: 4096000 | consumed tokens: 8388608000 | elapsed time per iteration (s): 0.30 | learning rate: 4.844E-05 | global batch size: 256 | lm loss: 3.371337E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.101 | TFLOPs: 30.32 | 0: steps: 16000 loss: 3.3591 iter time (s): 0.294 samples/sec: 871.885 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 16000 | lm loss value: 3.983759E+00 | lm loss PPL: 5.371858E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 16000 to checkpoints_146m14b100mdedup 0: [2023-03-17 01:15:17,908] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step16000 is begin to save! 0: [2023-03-17 01:15:17,911] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:15:17,999] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:15:17,999] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:15:18,014] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:15:18,015] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:15:18,029] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:15:18,030] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:15:18,044] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:15:18,045] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:15:18,059] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:15:18,060] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:15:18,074] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:15:18,074] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:15:18,089] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:15:18,089] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:15:18,104] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:15:18,104] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:15:18,119] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:15:18,119] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:15:18,134] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:15:18,134] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:15:18,149] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:15:18,149] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:15:18,164] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:15:18,164] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:15:18,227] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:15:18,228] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:15:18,243] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:15:18,244] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:15:18,260] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:15:18,260] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:15:18,275] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:15:18,276] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:15:18,277] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:15:18,277] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step16000/mp_rank_00_model_states.pt 0: [2023-03-17 01:15:18,277] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/mp_rank_00_model_states.pt... 0: [2023-03-17 01:15:18,280] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/mp_rank_00_model_states.pt. 0: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:15:18,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:15:18,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:15:18,329] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:15:18,329] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 01:15:18,329] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 0: [2023-03-17 01:15:18,330] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,330] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:15:18,328] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 01:15:18,329] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 6: [2023-03-17 01:15:18,331] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:15:18,331] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 01:15:18,331] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 5: [2023-03-17 01:15:18,331] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:15:18,332] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 01:15:18,332] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 0: [2023-03-17 01:15:18,332] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:15:18,332] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:15:18,332] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,332] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 6: [2023-03-17 01:15:18,332] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 01:15:18,332] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 3: [2023-03-17 01:15:18,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:15:18,333] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 01:15:18,333] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 6: [2023-03-17 01:15:18,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:15:18,333] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 01:15:18,333] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 6: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 3: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 3: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 1: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 5: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 4: [2023-03-17 01:15:18,330] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 7: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,330] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 4: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 7: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 7: [2023-03-17 01:15:18,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 4: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: [2023-03-17 01:15:18,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 0: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 4: [2023-03-17 01:15:18,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:15:18,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 0: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:15:18,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 0: [2023-03-17 01:15:18,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 5: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 4: [2023-03-17 01:15:18,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 01:15:18,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 0: [2023-03-17 01:15:18,336] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:15:18,337] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 5: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:15:18,337] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 2: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,337] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 3: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 3: [2023-03-17 01:15:18,337] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 1: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,337] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 6: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,337] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 0: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:15:18,338] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 1: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,338] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:15:18,338] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 01:15:18,338] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 1: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 1: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 0: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:15:18,338] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 5: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:15:18,339] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 01:15:18,339] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 3: [2023-03-17 01:15:18,339] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:15:18,340] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 5: [2023-03-17 01:15:18,339] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:15:18,340] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 5: [2023-03-17 01:15:18,340] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 3: [2023-03-17 01:15:18,340] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:15:18,340] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 3: [2023-03-17 01:15:18,340] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 01:15:18,340] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 3: [2023-03-17 01:15:18,340] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:15:18,340] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 2: [2023-03-17 01:15:18,337] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 2: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:15:18,338] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 3: [2023-03-17 01:15:18,340] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 2: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 2: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:15:18,338] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:15:18,338] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 2: [2023-03-17 01:15:18,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 4: [2023-03-17 01:15:18,341] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,341] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 01:15:18,341] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 3: [2023-03-17 01:15:18,341] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:15:18,341] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:15:18,341] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 2: [2023-03-17 01:15:18,341] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 3: [2023-03-17 01:15:18,341] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 2: [2023-03-17 01:15:18,341] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 1: [2023-03-17 01:15:18,341] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,342] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 1: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:15:18,342] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 2: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:15:18,342] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 2: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:15:18,342] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 5: [2023-03-17 01:15:18,344] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:15:18,344] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 01:15:18,344] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 5: [2023-03-17 01:15:18,344] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:15:18,344] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 01:15:18,344] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 6: [2023-03-17 01:15:18,337] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:15:18,337] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 6: [2023-03-17 01:15:18,346] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:15:18,346] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 01:15:18,346] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 6: [2023-03-17 01:15:18,347] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:15:18,348] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 01:15:18,348] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 4: [2023-03-17 01:15:18,342] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 4: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:15:18,342] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 0: [2023-03-17 01:15:18,349] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:15:18,349] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: [2023-03-17 01:15:18,339] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 0: successfully saved checkpoint at iteration 16000 to checkpoints_146m14b100mdedup 7: [2023-03-17 01:15:18,339] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:15:18,339] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: [2023-03-17 01:15:18,339] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:15:18,339] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 01:15:18,339] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:15:18,342] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step16000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:15:18,342] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step16000 is ready now! 7: time (ms) | save-checkpoint: 447.99 7: iteration 16010/ 21553 | consumed samples: 4098560 | consumed tokens: 8393850880 | elapsed time per iteration (s): 0.35 | learning rate: 4.835E-05 | global batch size: 256 | lm loss: 3.376476E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 726.935 | TFLOPs: 25.45 | 7: iteration 16020/ 21553 | consumed samples: 4101120 | consumed tokens: 8399093760 | elapsed time per iteration (s): 0.30 | learning rate: 4.825E-05 | global batch size: 256 | lm loss: 3.370527E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.202 | TFLOPs: 30.36 | 7: iteration 16030/ 21553 | consumed samples: 4103680 | consumed tokens: 8404336640 | elapsed time per iteration (s): 0.30 | learning rate: 4.815E-05 | global batch size: 256 | lm loss: 3.372269E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.027 | TFLOPs: 30.35 | 7: iteration 16040/ 21553 | consumed samples: 4106240 | consumed tokens: 8409579520 | elapsed time per iteration (s): 0.30 | learning rate: 4.806E-05 | global batch size: 256 | lm loss: 3.378947E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.148 | TFLOPs: 30.36 | 7: iteration 16050/ 21553 | consumed samples: 4108800 | consumed tokens: 8414822400 | elapsed time per iteration (s): 0.30 | learning rate: 4.796E-05 | global batch size: 256 | lm loss: 3.375847E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.962 | TFLOPs: 30.35 | 7: iteration 16060/ 21553 | consumed samples: 4111360 | consumed tokens: 8420065280 | elapsed time per iteration (s): 0.30 | learning rate: 4.786E-05 | global batch size: 256 | lm loss: 3.375196E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.559 | TFLOPs: 30.34 | 7: iteration 16070/ 21553 | consumed samples: 4113920 | consumed tokens: 8425308160 | elapsed time per iteration (s): 0.30 | learning rate: 4.777E-05 | global batch size: 256 | lm loss: 3.367734E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.617 | TFLOPs: 30.30 | 7: iteration 16080/ 21553 | consumed samples: 4116480 | consumed tokens: 8430551040 | elapsed time per iteration (s): 0.30 | learning rate: 4.767E-05 | global batch size: 256 | lm loss: 3.370429E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.681 | TFLOPs: 30.31 | 7: iteration 16090/ 21553 | consumed samples: 4119040 | consumed tokens: 8435793920 | elapsed time per iteration (s): 0.30 | learning rate: 4.758E-05 | global batch size: 256 | lm loss: 3.371511E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.867 | TFLOPs: 30.28 | 7: iteration 16100/ 21553 | consumed samples: 4121600 | consumed tokens: 8441036800 | elapsed time per iteration (s): 0.30 | learning rate: 4.748E-05 | global batch size: 256 | lm loss: 3.374693E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.565 | TFLOPs: 30.30 | 7: iteration 16110/ 21553 | consumed samples: 4124160 | consumed tokens: 8446279680 | elapsed time per iteration (s): 0.30 | learning rate: 4.739E-05 | global batch size: 256 | lm loss: 3.370216E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.988 | TFLOPs: 30.28 | 7: iteration 16120/ 21553 | consumed samples: 4126720 | consumed tokens: 8451522560 | elapsed time per iteration (s): 0.30 | learning rate: 4.729E-05 | global batch size: 256 | lm loss: 3.374983E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.116 | TFLOPs: 30.29 | 7: iteration 16130/ 21553 | consumed samples: 4129280 | consumed tokens: 8456765440 | elapsed time per iteration (s): 0.30 | learning rate: 4.720E-05 | global batch size: 256 | lm loss: 3.369020E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.147 | TFLOPs: 30.29 | 7: iteration 16140/ 21553 | consumed samples: 4131840 | consumed tokens: 8462008320 | elapsed time per iteration (s): 0.30 | learning rate: 4.710E-05 | global batch size: 256 | lm loss: 3.372844E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.798 | TFLOPs: 30.27 | 7: iteration 16150/ 21553 | consumed samples: 4134400 | consumed tokens: 8467251200 | elapsed time per iteration (s): 0.30 | learning rate: 4.701E-05 | global batch size: 256 | lm loss: 3.370589E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.307 | TFLOPs: 30.29 | 7: iteration 16160/ 21553 | consumed samples: 4136960 | consumed tokens: 8472494080 | elapsed time per iteration (s): 0.30 | learning rate: 4.691E-05 | global batch size: 256 | lm loss: 3.375021E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.941 | TFLOPs: 30.28 | 7: iteration 16170/ 21553 | consumed samples: 4139520 | consumed tokens: 8477736960 | elapsed time per iteration (s): 0.30 | learning rate: 4.682E-05 | global batch size: 256 | lm loss: 3.370562E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.872 | TFLOPs: 30.31 | 7: iteration 16180/ 21553 | consumed samples: 4142080 | consumed tokens: 8482979840 | elapsed time per iteration (s): 0.30 | learning rate: 4.672E-05 | global batch size: 256 | lm loss: 3.367986E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.171 | TFLOPs: 30.32 | 7: iteration 16190/ 21553 | consumed samples: 4144640 | consumed tokens: 8488222720 | elapsed time per iteration (s): 0.30 | learning rate: 4.663E-05 | global batch size: 256 | lm loss: 3.374786E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.185 | TFLOPs: 30.29 | 7: iteration 16200/ 21553 | consumed samples: 4147200 | consumed tokens: 8493465600 | elapsed time per iteration (s): 0.30 | learning rate: 4.654E-05 | global batch size: 256 | lm loss: 3.369537E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.945 | TFLOPs: 30.31 | 7: iteration 16210/ 21553 | consumed samples: 4149760 | consumed tokens: 8498708480 | elapsed time per iteration (s): 0.30 | learning rate: 4.644E-05 | global batch size: 256 | lm loss: 3.366444E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.230 | TFLOPs: 30.32 | 7: iteration 16220/ 21553 | consumed samples: 4152320 | consumed tokens: 8503951360 | elapsed time per iteration (s): 0.30 | learning rate: 4.635E-05 | global batch size: 256 | lm loss: 3.375832E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.839 | TFLOPs: 30.31 | 7: iteration 16230/ 21553 | consumed samples: 4154880 | consumed tokens: 8509194240 | elapsed time per iteration (s): 0.30 | learning rate: 4.625E-05 | global batch size: 256 | lm loss: 3.367007E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.990 | TFLOPs: 30.32 | 7: iteration 16240/ 21553 | consumed samples: 4157440 | consumed tokens: 8514437120 | elapsed time per iteration (s): 0.30 | learning rate: 4.616E-05 | global batch size: 256 | lm loss: 3.368038E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.996 | TFLOPs: 30.28 | 7: iteration 16250/ 21553 | consumed samples: 4160000 | consumed tokens: 8519680000 | elapsed time per iteration (s): 0.30 | learning rate: 4.607E-05 | global batch size: 256 | lm loss: 3.370967E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.737 | TFLOPs: 30.24 | 7: iteration 16260/ 21553 | consumed samples: 4162560 | consumed tokens: 8524922880 | elapsed time per iteration (s): 0.30 | learning rate: 4.597E-05 | global batch size: 256 | lm loss: 3.372912E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.900 | TFLOPs: 30.35 | 7: iteration 16270/ 21553 | consumed samples: 4165120 | consumed tokens: 8530165760 | elapsed time per iteration (s): 0.30 | learning rate: 4.588E-05 | global batch size: 256 | lm loss: 3.372268E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.634 | TFLOPs: 30.34 | 7: iteration 16280/ 21553 | consumed samples: 4167680 | consumed tokens: 8535408640 | elapsed time per iteration (s): 0.30 | learning rate: 4.579E-05 | global batch size: 256 | lm loss: 3.365312E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.246 | TFLOPs: 30.32 | 7: iteration 16290/ 21553 | consumed samples: 4170240 | consumed tokens: 8540651520 | elapsed time per iteration (s): 0.30 | learning rate: 4.570E-05 | global batch size: 256 | lm loss: 3.371119E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.490 | TFLOPs: 30.33 | 7: iteration 16300/ 21553 | consumed samples: 4172800 | consumed tokens: 8545894400 | elapsed time per iteration (s): 0.30 | learning rate: 4.560E-05 | global batch size: 256 | lm loss: 3.372179E+00 | grad norm: 0.404 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.949 | TFLOPs: 30.35 | 7: iteration 16310/ 21553 | consumed samples: 4175360 | consumed tokens: 8551137280 | elapsed time per iteration (s): 0.30 | learning rate: 4.551E-05 | global batch size: 256 | lm loss: 3.365465E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.519 | TFLOPs: 30.37 | 7: iteration 16320/ 21553 | consumed samples: 4177920 | consumed tokens: 8556380160 | elapsed time per iteration (s): 0.30 | learning rate: 4.542E-05 | global batch size: 256 | lm loss: 3.366151E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.944 | TFLOPs: 30.35 | 7: iteration 16330/ 21553 | consumed samples: 4180480 | consumed tokens: 8561623040 | elapsed time per iteration (s): 0.30 | learning rate: 4.533E-05 | global batch size: 256 | lm loss: 3.366971E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.591 | TFLOPs: 30.37 | 7: iteration 16340/ 21553 | consumed samples: 4183040 | consumed tokens: 8566865920 | elapsed time per iteration (s): 0.30 | learning rate: 4.523E-05 | global batch size: 256 | lm loss: 3.361849E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.011 | TFLOPs: 30.35 | 7: iteration 16350/ 21553 | consumed samples: 4185600 | consumed tokens: 8572108800 | elapsed time per iteration (s): 0.30 | learning rate: 4.514E-05 | global batch size: 256 | lm loss: 3.363437E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.554 | TFLOPs: 30.34 | 7: iteration 16360/ 21553 | consumed samples: 4188160 | consumed tokens: 8577351680 | elapsed time per iteration (s): 0.30 | learning rate: 4.505E-05 | global batch size: 256 | lm loss: 3.368825E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.337 | TFLOPs: 30.33 | 7: iteration 16370/ 21553 | consumed samples: 4190720 | consumed tokens: 8582594560 | elapsed time per iteration (s): 0.30 | learning rate: 4.496E-05 | global batch size: 256 | lm loss: 3.370097E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.701 | TFLOPs: 30.34 | 7: iteration 16380/ 21553 | consumed samples: 4193280 | consumed tokens: 8587837440 | elapsed time per iteration (s): 0.30 | learning rate: 4.487E-05 | global batch size: 256 | lm loss: 3.363628E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.545 | TFLOPs: 30.34 | 7: iteration 16390/ 21553 | consumed samples: 4195840 | consumed tokens: 8593080320 | elapsed time per iteration (s): 0.30 | learning rate: 4.478E-05 | global batch size: 256 | lm loss: 3.366528E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.298 | TFLOPs: 30.33 | 7: iteration 16400/ 21553 | consumed samples: 4198400 | consumed tokens: 8598323200 | elapsed time per iteration (s): 0.30 | learning rate: 4.468E-05 | global batch size: 256 | lm loss: 3.361427E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.817 | TFLOPs: 30.31 | 7: iteration 16410/ 21553 | consumed samples: 4200960 | consumed tokens: 8603566080 | elapsed time per iteration (s): 0.30 | learning rate: 4.459E-05 | global batch size: 256 | lm loss: 3.366244E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.812 | TFLOPs: 30.34 | 7: iteration 16420/ 21553 | consumed samples: 4203520 | consumed tokens: 8608808960 | elapsed time per iteration (s): 0.30 | learning rate: 4.450E-05 | global batch size: 256 | lm loss: 3.370480E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.282 | TFLOPs: 30.36 | 7: iteration 16430/ 21553 | consumed samples: 4206080 | consumed tokens: 8614051840 | elapsed time per iteration (s): 0.30 | learning rate: 4.441E-05 | global batch size: 256 | lm loss: 3.371040E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.464 | TFLOPs: 30.37 | 7: iteration 16440/ 21553 | consumed samples: 4208640 | consumed tokens: 8619294720 | elapsed time per iteration (s): 0.30 | learning rate: 4.432E-05 | global batch size: 256 | lm loss: 3.368239E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.226 | TFLOPs: 30.36 | 7: iteration 16450/ 21553 | consumed samples: 4211200 | consumed tokens: 8624537600 | elapsed time per iteration (s): 0.30 | learning rate: 4.423E-05 | global batch size: 256 | lm loss: 3.368532E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.623 | TFLOPs: 30.37 | 7: iteration 16460/ 21553 | consumed samples: 4213760 | consumed tokens: 8629780480 | elapsed time per iteration (s): 0.30 | learning rate: 4.414E-05 | global batch size: 256 | lm loss: 3.368954E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.945 | TFLOPs: 30.35 | 7: iteration 16470/ 21553 | consumed samples: 4216320 | consumed tokens: 8635023360 | elapsed time per iteration (s): 0.30 | learning rate: 4.405E-05 | global batch size: 256 | lm loss: 3.367126E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.731 | TFLOPs: 30.34 | 7: iteration 16480/ 21553 | consumed samples: 4218880 | consumed tokens: 8640266240 | elapsed time per iteration (s): 0.30 | learning rate: 4.396E-05 | global batch size: 256 | lm loss: 3.359343E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.500 | TFLOPs: 30.33 | 7: iteration 16490/ 21553 | consumed samples: 4221440 | consumed tokens: 8645509120 | elapsed time per iteration (s): 0.30 | learning rate: 4.387E-05 | global batch size: 256 | lm loss: 3.366915E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.124 | TFLOPs: 30.32 | 7: iteration 16500/ 21553 | consumed samples: 4224000 | consumed tokens: 8650752000 | elapsed time per iteration (s): 0.30 | learning rate: 4.378E-05 | global batch size: 256 | lm loss: 3.369249E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.273 | TFLOPs: 30.33 | 7: iteration 16510/ 21553 | consumed samples: 4226560 | consumed tokens: 8655994880 | elapsed time per iteration (s): 0.30 | learning rate: 4.369E-05 | global batch size: 256 | lm loss: 3.370159E+00 | grad norm: 0.456 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.497 | TFLOPs: 30.33 | 7: iteration 16520/ 21553 | consumed samples: 4229120 | consumed tokens: 8661237760 | elapsed time per iteration (s): 0.30 | learning rate: 4.360E-05 | global batch size: 256 | lm loss: 3.363383E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.957 | TFLOPs: 30.35 | 7: iteration 16530/ 21553 | consumed samples: 4231680 | consumed tokens: 8666480640 | elapsed time per iteration (s): 0.30 | learning rate: 4.351E-05 | global batch size: 256 | lm loss: 3.366761E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 852.202 | TFLOPs: 29.83 | 7: iteration 16540/ 21553 | consumed samples: 4234240 | consumed tokens: 8671723520 | elapsed time per iteration (s): 0.30 | learning rate: 4.342E-05 | global batch size: 256 | lm loss: 3.369635E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.543 | TFLOPs: 30.37 | 7: iteration 16550/ 21553 | consumed samples: 4236800 | consumed tokens: 8676966400 | elapsed time per iteration (s): 0.30 | learning rate: 4.333E-05 | global batch size: 256 | lm loss: 3.368784E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.994 | TFLOPs: 30.35 | 7: iteration 16560/ 21553 | consumed samples: 4239360 | consumed tokens: 8682209280 | elapsed time per iteration (s): 0.30 | learning rate: 4.324E-05 | global batch size: 256 | lm loss: 3.364522E+00 | grad norm: 0.466 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.336 | TFLOPs: 30.36 | 7: iteration 16570/ 21553 | consumed samples: 4241920 | consumed tokens: 8687452160 | elapsed time per iteration (s): 0.30 | learning rate: 4.315E-05 | global batch size: 256 | lm loss: 3.358709E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.529 | TFLOPs: 30.37 | 7: iteration 16580/ 21553 | consumed samples: 4244480 | consumed tokens: 8692695040 | elapsed time per iteration (s): 0.30 | learning rate: 4.307E-05 | global batch size: 256 | lm loss: 3.364380E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.341 | TFLOPs: 30.36 | 7: iteration 16590/ 21553 | consumed samples: 4247040 | consumed tokens: 8697937920 | elapsed time per iteration (s): 0.30 | learning rate: 4.298E-05 | global batch size: 256 | lm loss: 3.367227E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.769 | TFLOPs: 30.34 | 7: iteration 16600/ 21553 | consumed samples: 4249600 | consumed tokens: 8703180800 | elapsed time per iteration (s): 0.30 | learning rate: 4.289E-05 | global batch size: 256 | lm loss: 3.359718E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.155 | TFLOPs: 30.36 | 7: iteration 16610/ 21553 | consumed samples: 4252160 | consumed tokens: 8708423680 | elapsed time per iteration (s): 0.30 | learning rate: 4.280E-05 | global batch size: 256 | lm loss: 3.362229E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.813 | TFLOPs: 30.34 | 7: iteration 16620/ 21553 | consumed samples: 4254720 | consumed tokens: 8713666560 | elapsed time per iteration (s): 0.30 | learning rate: 4.271E-05 | global batch size: 256 | lm loss: 3.367107E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.025 | TFLOPs: 30.35 | 7: iteration 16630/ 21553 | consumed samples: 4257280 | consumed tokens: 8718909440 | elapsed time per iteration (s): 0.30 | learning rate: 4.263E-05 | global batch size: 256 | lm loss: 3.365807E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.782 | TFLOPs: 30.34 | 7: iteration 16640/ 21553 | consumed samples: 4259840 | consumed tokens: 8724152320 | elapsed time per iteration (s): 0.30 | learning rate: 4.254E-05 | global batch size: 256 | lm loss: 3.364229E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.751 | TFLOPs: 30.34 | 7: iteration 16650/ 21553 | consumed samples: 4262400 | consumed tokens: 8729395200 | elapsed time per iteration (s): 0.30 | learning rate: 4.245E-05 | global batch size: 256 | lm loss: 3.365099E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.017 | TFLOPs: 30.35 | 7: iteration 16660/ 21553 | consumed samples: 4264960 | consumed tokens: 8734638080 | elapsed time per iteration (s): 0.29 | learning rate: 4.236E-05 | global batch size: 256 | lm loss: 3.364904E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.877 | TFLOPs: 30.38 | 7: iteration 16670/ 21553 | consumed samples: 4267520 | consumed tokens: 8739880960 | elapsed time per iteration (s): 0.30 | learning rate: 4.227E-05 | global batch size: 256 | lm loss: 3.360597E+00 | grad norm: 0.400 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.399 | TFLOPs: 30.37 | 7: iteration 16680/ 21553 | consumed samples: 4270080 | consumed tokens: 8745123840 | elapsed time per iteration (s): 0.30 | learning rate: 4.219E-05 | global batch size: 256 | lm loss: 3.363752E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.666 | TFLOPs: 30.37 | 7: iteration 16690/ 21553 | consumed samples: 4272640 | consumed tokens: 8750366720 | elapsed time per iteration (s): 0.29 | learning rate: 4.210E-05 | global batch size: 256 | lm loss: 3.363624E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.815 | TFLOPs: 30.38 | 7: iteration 16700/ 21553 | consumed samples: 4275200 | consumed tokens: 8755609600 | elapsed time per iteration (s): 0.30 | learning rate: 4.201E-05 | global batch size: 256 | lm loss: 3.370544E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.628 | TFLOPs: 30.37 | 7: iteration 16710/ 21553 | consumed samples: 4277760 | consumed tokens: 8760852480 | elapsed time per iteration (s): 0.30 | learning rate: 4.193E-05 | global batch size: 256 | lm loss: 3.363073E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.622 | TFLOPs: 30.37 | 7: iteration 16720/ 21553 | consumed samples: 4280320 | consumed tokens: 8766095360 | elapsed time per iteration (s): 0.30 | learning rate: 4.184E-05 | global batch size: 256 | lm loss: 3.362741E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.540 | TFLOPs: 30.37 | 7: iteration 16730/ 21553 | consumed samples: 4282880 | consumed tokens: 8771338240 | elapsed time per iteration (s): 0.30 | learning rate: 4.175E-05 | global batch size: 256 | lm loss: 3.362504E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.684 | TFLOPs: 30.38 | 7: iteration 16740/ 21553 | consumed samples: 4285440 | consumed tokens: 8776581120 | elapsed time per iteration (s): 0.30 | learning rate: 4.167E-05 | global batch size: 256 | lm loss: 3.360112E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.597 | TFLOPs: 30.37 | 7: iteration 16750/ 21553 | consumed samples: 4288000 | consumed tokens: 8781824000 | elapsed time per iteration (s): 0.29 | learning rate: 4.158E-05 | global batch size: 256 | lm loss: 3.358245E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.906 | TFLOPs: 30.38 | 7: iteration 16760/ 21553 | consumed samples: 4290560 | consumed tokens: 8787066880 | elapsed time per iteration (s): 0.30 | learning rate: 4.150E-05 | global batch size: 256 | lm loss: 3.368617E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.363 | TFLOPs: 30.36 | 7: iteration 16770/ 21553 | consumed samples: 4293120 | consumed tokens: 8792309760 | elapsed time per iteration (s): 0.29 | learning rate: 4.141E-05 | global batch size: 256 | lm loss: 3.358574E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.907 | TFLOPs: 30.38 | 7: iteration 16780/ 21553 | consumed samples: 4295680 | consumed tokens: 8797552640 | elapsed time per iteration (s): 0.30 | learning rate: 4.132E-05 | global batch size: 256 | lm loss: 3.360117E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.670 | TFLOPs: 30.37 | 7: iteration 16790/ 21553 | consumed samples: 4298240 | consumed tokens: 8802795520 | elapsed time per iteration (s): 0.29 | learning rate: 4.124E-05 | global batch size: 256 | lm loss: 3.358748E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.916 | TFLOPs: 30.38 | 7: iteration 16800/ 21553 | consumed samples: 4300800 | consumed tokens: 8808038400 | elapsed time per iteration (s): 0.30 | learning rate: 4.115E-05 | global batch size: 256 | lm loss: 3.368475E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.875 | TFLOPs: 30.35 | 7: iteration 16810/ 21553 | consumed samples: 4303360 | consumed tokens: 8813281280 | elapsed time per iteration (s): 0.30 | learning rate: 4.107E-05 | global batch size: 256 | lm loss: 3.359298E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.932 | TFLOPs: 30.35 | 7: iteration 16820/ 21553 | consumed samples: 4305920 | consumed tokens: 8818524160 | elapsed time per iteration (s): 0.30 | learning rate: 4.098E-05 | global batch size: 256 | lm loss: 3.368977E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.335 | TFLOPs: 30.36 | 7: iteration 16830/ 21553 | consumed samples: 4308480 | consumed tokens: 8823767040 | elapsed time per iteration (s): 0.30 | learning rate: 4.090E-05 | global batch size: 256 | lm loss: 3.361559E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.591 | TFLOPs: 30.37 | 7: iteration 16840/ 21553 | consumed samples: 4311040 | consumed tokens: 8829009920 | elapsed time per iteration (s): 0.29 | learning rate: 4.081E-05 | global batch size: 256 | lm loss: 3.352858E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.019 | TFLOPs: 30.39 | 7: iteration 16850/ 21553 | consumed samples: 4313600 | consumed tokens: 8834252800 | elapsed time per iteration (s): 0.30 | learning rate: 4.073E-05 | global batch size: 256 | lm loss: 3.364554E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.704 | TFLOPs: 30.38 | 7: iteration 16860/ 21553 | consumed samples: 4316160 | consumed tokens: 8839495680 | elapsed time per iteration (s): 0.30 | learning rate: 4.064E-05 | global batch size: 256 | lm loss: 3.366750E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.540 | TFLOPs: 30.37 | 7: iteration 16870/ 21553 | consumed samples: 4318720 | consumed tokens: 8844738560 | elapsed time per iteration (s): 0.30 | learning rate: 4.056E-05 | global batch size: 256 | lm loss: 3.362772E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 849.944 | TFLOPs: 29.75 | 7: iteration 16880/ 21553 | consumed samples: 4321280 | consumed tokens: 8849981440 | elapsed time per iteration (s): 0.29 | learning rate: 4.047E-05 | global batch size: 256 | lm loss: 3.362191E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.884 | TFLOPs: 30.38 | 7: iteration 16890/ 21553 | consumed samples: 4323840 | consumed tokens: 8855224320 | elapsed time per iteration (s): 0.30 | learning rate: 4.039E-05 | global batch size: 256 | lm loss: 3.356000E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.668 | TFLOPs: 30.37 | 7: iteration 16900/ 21553 | consumed samples: 4326400 | consumed tokens: 8860467200 | elapsed time per iteration (s): 0.29 | learning rate: 4.031E-05 | global batch size: 256 | lm loss: 3.353886E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.939 | TFLOPs: 30.38 | 7: iteration 16910/ 21553 | consumed samples: 4328960 | consumed tokens: 8865710080 | elapsed time per iteration (s): 0.29 | learning rate: 4.022E-05 | global batch size: 256 | lm loss: 3.362918E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.925 | TFLOPs: 30.38 | 7: iteration 16920/ 21553 | consumed samples: 4331520 | consumed tokens: 8870952960 | elapsed time per iteration (s): 0.29 | learning rate: 4.014E-05 | global batch size: 256 | lm loss: 3.358204E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.973 | TFLOPs: 30.39 | 7: iteration 16930/ 21553 | consumed samples: 4334080 | consumed tokens: 8876195840 | elapsed time per iteration (s): 0.29 | learning rate: 4.006E-05 | global batch size: 256 | lm loss: 3.358547E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.823 | TFLOPs: 30.38 | 7: iteration 16940/ 21553 | consumed samples: 4336640 | consumed tokens: 8881438720 | elapsed time per iteration (s): 0.29 | learning rate: 3.997E-05 | global batch size: 256 | lm loss: 3.357346E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.052 | TFLOPs: 30.39 | 7: iteration 16950/ 21553 | consumed samples: 4339200 | consumed tokens: 8886681600 | elapsed time per iteration (s): 0.30 | learning rate: 3.989E-05 | global batch size: 256 | lm loss: 3.350766E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.629 | TFLOPs: 30.37 | 7: iteration 16960/ 21553 | consumed samples: 4341760 | consumed tokens: 8891924480 | elapsed time per iteration (s): 0.30 | learning rate: 3.981E-05 | global batch size: 256 | lm loss: 3.360491E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.547 | TFLOPs: 30.37 | 7: iteration 16970/ 21553 | consumed samples: 4344320 | consumed tokens: 8897167360 | elapsed time per iteration (s): 0.29 | learning rate: 3.972E-05 | global batch size: 256 | lm loss: 3.350871E+00 | grad norm: 0.399 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 868.394 | TFLOPs: 30.40 | 7: iteration 16980/ 21553 | consumed samples: 4346880 | consumed tokens: 8902410240 | elapsed time per iteration (s): 0.30 | learning rate: 3.964E-05 | global batch size: 256 | lm loss: 3.357521E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.780 | TFLOPs: 30.38 | 7: iteration 16990/ 21553 | consumed samples: 4349440 | consumed tokens: 8907653120 | elapsed time per iteration (s): 0.30 | learning rate: 3.956E-05 | global batch size: 256 | lm loss: 3.359394E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.104 | TFLOPs: 30.35 | 7: iteration 17000/ 21553 | consumed samples: 4352000 | consumed tokens: 8912896000 | elapsed time per iteration (s): 0.30 | learning rate: 3.948E-05 | global batch size: 256 | lm loss: 3.361385E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.012 | TFLOPs: 30.35 | 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 17000 | lm loss value: 3.921767E+00 | lm loss PPL: 5.048959E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 17000 to checkpoints_146m14b100mdedup 0: [2023-03-17 01:20:13,888] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step17000 is begin to save! 0: [2023-03-17 01:20:13,892] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:20:13,979] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:20:13,979] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:20:13,994] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:20:13,995] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:20:14,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:20:14,010] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:20:14,024] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:20:14,025] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:20:14,039] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:20:14,040] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:20:14,054] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:20:14,055] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:20:14,069] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:20:14,070] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:20:14,084] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:20:14,085] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:20:14,099] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:20:14,100] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:20:14,114] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:20:14,115] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:20:14,129] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:20:14,130] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:20:14,144] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:20:14,145] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:20:14,159] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:20:14,160] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:20:14,174] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:20:14,175] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:20:14,190] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:20:14,190] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:20:14,205] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:20:14,205] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:20:14,206] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:20:14,207] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step17000/mp_rank_00_model_states.pt 0: [2023-03-17 01:20:14,207] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/mp_rank_00_model_states.pt... 0: [2023-03-17 01:20:14,209] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/mp_rank_00_model_states.pt. 0: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:20:14,226] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:20:14,261] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:20:14,261] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 2: [2023-03-17 01:20:14,261] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:20:14,261] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,261] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 01:20:14,261] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 5: [2023-03-17 01:20:14,262] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,262] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 01:20:14,262] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 5: [2023-03-17 01:20:14,262] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:20:14,262] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,262] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 3: [2023-03-17 01:20:14,262] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 5: [2023-03-17 01:20:14,262] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 3: [2023-03-17 01:20:14,262] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 0: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:20:14,263] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,264] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 0: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:20:14,264] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 0: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:20:14,264] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 01:20:14,264] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 0: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 1: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:20:14,265] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,264] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 1: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 7: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:20:14,264] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 7: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:20:14,265] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 1: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 1: [2023-03-17 01:20:14,265] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 5: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 3: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,265] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 3: [2023-03-17 01:20:14,265] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 01:20:14,265] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,263] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,263] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:20:14,263] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,263] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:20:14,264] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:20:14,264] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 3: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:20:14,266] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 6: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:20:14,266] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:20:14,266] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 6: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 3: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:20:14,267] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 1: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:20:14,267] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:20:14,267] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:20:14,267] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 01:20:14,267] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 1: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 1: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 1: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 5: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,266] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 7: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 5: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,266] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,266] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 7: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 7: [2023-03-17 01:20:14,267] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,267] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 5: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 1: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 6: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:20:14,269] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 6: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:20:14,270] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 01:20:14,270] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 6: [2023-03-17 01:20:14,270] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:20:14,270] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 01:20:14,270] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 6: [2023-03-17 01:20:14,270] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:20:14,270] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 01:20:14,270] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,268] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 01:20:14,268] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:20:14,269] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:20:14,269] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 2: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 3: [2023-03-17 01:20:14,272] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:20:14,272] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:20:14,272] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 1: [2023-03-17 01:20:14,272] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 3: [2023-03-17 01:20:14,272] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 1: [2023-03-17 01:20:14,272] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 7: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:20:14,269] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 7: [2023-03-17 01:20:14,269] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:20:14,270] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:20:14,270] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 4: [2023-03-17 01:20:14,274] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:20:14,275] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 01:20:14,275] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 3: [2023-03-17 01:20:14,275] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:20:14,275] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:20:14,275] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 01:20:14,275] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 01:20:14,275] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 3: [2023-03-17 01:20:14,275] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 6: [2023-03-17 01:20:14,276] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:20:14,276] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:20:14,276] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:20:14,276] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:20:14,276] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 01:20:14,276] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 0: [2023-03-17 01:20:14,276] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:20:14,276] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 6: [2023-03-17 01:20:14,276] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 6: [2023-03-17 01:20:14,276] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 0: [2023-03-17 01:20:14,276] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 0: [2023-03-17 01:20:14,276] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 0: [2023-03-17 01:20:14,280] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:20:14,280] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:20:14,280] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 01:20:14,280] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 0: [2023-03-17 01:20:14,290] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step17000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:20:14,290] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step17000 is ready now! 0: successfully saved checkpoint at iteration 17000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 407.31 7: iteration 17010/ 21553 | consumed samples: 4354560 | consumed tokens: 8918138880 | elapsed time per iteration (s): 0.35 | learning rate: 3.939E-05 | global batch size: 256 | lm loss: 3.355069E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 741.727 | TFLOPs: 25.97 | 7: iteration 17020/ 21553 | consumed samples: 4357120 | consumed tokens: 8923381760 | elapsed time per iteration (s): 0.30 | learning rate: 3.931E-05 | global batch size: 256 | lm loss: 3.356895E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.397 | TFLOPs: 30.37 | 7: iteration 17030/ 21553 | consumed samples: 4359680 | consumed tokens: 8928624640 | elapsed time per iteration (s): 0.30 | learning rate: 3.923E-05 | global batch size: 256 | lm loss: 3.359767E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.068 | TFLOPs: 30.35 | 7: iteration 17040/ 21553 | consumed samples: 4362240 | consumed tokens: 8933867520 | elapsed time per iteration (s): 0.30 | learning rate: 3.915E-05 | global batch size: 256 | lm loss: 3.361795E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.340 | TFLOPs: 30.36 | 7: iteration 17050/ 21553 | consumed samples: 4364800 | consumed tokens: 8939110400 | elapsed time per iteration (s): 0.30 | learning rate: 3.907E-05 | global batch size: 256 | lm loss: 3.366360E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.721 | TFLOPs: 30.34 | 7: iteration 17060/ 21553 | consumed samples: 4367360 | consumed tokens: 8944353280 | elapsed time per iteration (s): 0.30 | learning rate: 3.898E-05 | global batch size: 256 | lm loss: 3.354126E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.228 | TFLOPs: 30.36 | 7: iteration 17070/ 21553 | consumed samples: 4369920 | consumed tokens: 8949596160 | elapsed time per iteration (s): 0.30 | learning rate: 3.890E-05 | global batch size: 256 | lm loss: 3.356380E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.700 | TFLOPs: 30.34 | 7: iteration 17080/ 21553 | consumed samples: 4372480 | consumed tokens: 8954839040 | elapsed time per iteration (s): 0.30 | learning rate: 3.882E-05 | global batch size: 256 | lm loss: 3.352411E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.980 | TFLOPs: 30.35 | 7: iteration 17090/ 21553 | consumed samples: 4375040 | consumed tokens: 8960081920 | elapsed time per iteration (s): 0.30 | learning rate: 3.874E-05 | global batch size: 256 | lm loss: 3.348621E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.525 | TFLOPs: 30.33 | 7: iteration 17100/ 21553 | consumed samples: 4377600 | consumed tokens: 8965324800 | elapsed time per iteration (s): 0.30 | learning rate: 3.866E-05 | global batch size: 256 | lm loss: 3.357038E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.824 | TFLOPs: 30.31 | 7: iteration 17110/ 21553 | consumed samples: 4380160 | consumed tokens: 8970567680 | elapsed time per iteration (s): 0.30 | learning rate: 3.858E-05 | global batch size: 256 | lm loss: 3.361568E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.483 | TFLOPs: 30.30 | 7: iteration 17120/ 21553 | consumed samples: 4382720 | consumed tokens: 8975810560 | elapsed time per iteration (s): 0.30 | learning rate: 3.850E-05 | global batch size: 256 | lm loss: 3.356417E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.266 | TFLOPs: 30.29 | 7: iteration 17130/ 21553 | consumed samples: 4385280 | consumed tokens: 8981053440 | elapsed time per iteration (s): 0.30 | learning rate: 3.842E-05 | global batch size: 256 | lm loss: 3.358520E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.290 | TFLOPs: 30.29 | 7: iteration 17140/ 21553 | consumed samples: 4387840 | consumed tokens: 8986296320 | elapsed time per iteration (s): 0.30 | learning rate: 3.834E-05 | global batch size: 256 | lm loss: 3.357104E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.787 | TFLOPs: 30.31 | 7: iteration 17150/ 21553 | consumed samples: 4390400 | consumed tokens: 8991539200 | elapsed time per iteration (s): 0.30 | learning rate: 3.826E-05 | global batch size: 256 | lm loss: 3.354381E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.851 | TFLOPs: 30.31 | 7: iteration 17160/ 21553 | consumed samples: 4392960 | consumed tokens: 8996782080 | elapsed time per iteration (s): 0.30 | learning rate: 3.818E-05 | global batch size: 256 | lm loss: 3.358735E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.030 | TFLOPs: 30.32 | 7: iteration 17170/ 21553 | consumed samples: 4395520 | consumed tokens: 9002024960 | elapsed time per iteration (s): 0.30 | learning rate: 3.810E-05 | global batch size: 256 | lm loss: 3.355994E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.918 | TFLOPs: 30.31 | 7: iteration 17180/ 21553 | consumed samples: 4398080 | consumed tokens: 9007267840 | elapsed time per iteration (s): 0.30 | learning rate: 3.802E-05 | global batch size: 256 | lm loss: 3.356252E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.882 | TFLOPs: 30.31 | 7: iteration 17190/ 21553 | consumed samples: 4400640 | consumed tokens: 9012510720 | elapsed time per iteration (s): 0.30 | learning rate: 3.794E-05 | global batch size: 256 | lm loss: 3.355983E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.375 | TFLOPs: 30.33 | 7: iteration 17200/ 21553 | consumed samples: 4403200 | consumed tokens: 9017753600 | elapsed time per iteration (s): 0.30 | learning rate: 3.786E-05 | global batch size: 256 | lm loss: 3.357305E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.468 | TFLOPs: 30.33 | 7: iteration 17210/ 21553 | consumed samples: 4405760 | consumed tokens: 9022996480 | elapsed time per iteration (s): 0.30 | learning rate: 3.778E-05 | global batch size: 256 | lm loss: 3.362539E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.004 | TFLOPs: 30.32 | 7: iteration 17220/ 21553 | consumed samples: 4408320 | consumed tokens: 9028239360 | elapsed time per iteration (s): 0.30 | learning rate: 3.770E-05 | global batch size: 256 | lm loss: 3.353963E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.150 | TFLOPs: 30.32 | 7: iteration 17230/ 21553 | consumed samples: 4410880 | consumed tokens: 9033482240 | elapsed time per iteration (s): 0.30 | learning rate: 3.762E-05 | global batch size: 256 | lm loss: 3.351159E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 860.918 | TFLOPs: 30.14 | 7: iteration 17240/ 21553 | consumed samples: 4413440 | consumed tokens: 9038725120 | elapsed time per iteration (s): 0.30 | learning rate: 3.754E-05 | global batch size: 256 | lm loss: 3.357328E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.455 | TFLOPs: 30.33 | 7: iteration 17250/ 21553 | consumed samples: 4416000 | consumed tokens: 9043968000 | elapsed time per iteration (s): 0.30 | learning rate: 3.747E-05 | global batch size: 256 | lm loss: 3.356288E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.767 | TFLOPs: 30.31 | 7: iteration 17260/ 21553 | consumed samples: 4418560 | consumed tokens: 9049210880 | elapsed time per iteration (s): 0.30 | learning rate: 3.739E-05 | global batch size: 256 | lm loss: 3.355772E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.591 | TFLOPs: 30.27 | 7: iteration 17270/ 21553 | consumed samples: 4421120 | consumed tokens: 9054453760 | elapsed time per iteration (s): 0.30 | learning rate: 3.731E-05 | global batch size: 256 | lm loss: 3.352027E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.610 | TFLOPs: 30.34 | 7: iteration 17280/ 21553 | consumed samples: 4423680 | consumed tokens: 9059696640 | elapsed time per iteration (s): 0.30 | learning rate: 3.723E-05 | global batch size: 256 | lm loss: 3.360961E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.850 | TFLOPs: 30.35 | 7: iteration 17290/ 21553 | consumed samples: 4426240 | consumed tokens: 9064939520 | elapsed time per iteration (s): 0.30 | learning rate: 3.715E-05 | global batch size: 256 | lm loss: 3.352528E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.609 | TFLOPs: 30.34 | 7: iteration 17300/ 21553 | consumed samples: 4428800 | consumed tokens: 9070182400 | elapsed time per iteration (s): 0.30 | learning rate: 3.708E-05 | global batch size: 256 | lm loss: 3.363584E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.291 | TFLOPs: 30.33 | 7: iteration 17310/ 21553 | consumed samples: 4431360 | consumed tokens: 9075425280 | elapsed time per iteration (s): 0.30 | learning rate: 3.700E-05 | global batch size: 256 | lm loss: 3.353723E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.366 | TFLOPs: 30.33 | 7: iteration 17320/ 21553 | consumed samples: 4433920 | consumed tokens: 9080668160 | elapsed time per iteration (s): 0.30 | learning rate: 3.692E-05 | global batch size: 256 | lm loss: 3.353242E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.683 | TFLOPs: 30.31 | 7: iteration 17330/ 21553 | consumed samples: 4436480 | consumed tokens: 9085911040 | elapsed time per iteration (s): 0.30 | learning rate: 3.684E-05 | global batch size: 256 | lm loss: 3.358519E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.355 | TFLOPs: 30.33 | 7: iteration 17340/ 21553 | consumed samples: 4439040 | consumed tokens: 9091153920 | elapsed time per iteration (s): 0.30 | learning rate: 3.677E-05 | global batch size: 256 | lm loss: 3.349429E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.129 | TFLOPs: 30.32 | 7: iteration 17350/ 21553 | consumed samples: 4441600 | consumed tokens: 9096396800 | elapsed time per iteration (s): 0.30 | learning rate: 3.669E-05 | global batch size: 256 | lm loss: 3.359388E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.316 | TFLOPs: 30.33 | 7: iteration 17360/ 21553 | consumed samples: 4444160 | consumed tokens: 9101639680 | elapsed time per iteration (s): 0.30 | learning rate: 3.661E-05 | global batch size: 256 | lm loss: 3.350161E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.458 | TFLOPs: 30.33 | 7: iteration 17370/ 21553 | consumed samples: 4446720 | consumed tokens: 9106882560 | elapsed time per iteration (s): 0.30 | learning rate: 3.654E-05 | global batch size: 256 | lm loss: 3.352243E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.400 | TFLOPs: 30.33 | 7: iteration 17380/ 21553 | consumed samples: 4449280 | consumed tokens: 9112125440 | elapsed time per iteration (s): 0.30 | learning rate: 3.646E-05 | global batch size: 256 | lm loss: 3.354044E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.574 | TFLOPs: 30.34 | 7: iteration 17390/ 21553 | consumed samples: 4451840 | consumed tokens: 9117368320 | elapsed time per iteration (s): 0.30 | learning rate: 3.638E-05 | global batch size: 256 | lm loss: 3.353032E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.643 | TFLOPs: 30.20 | 7: iteration 17400/ 21553 | consumed samples: 4454400 | consumed tokens: 9122611200 | elapsed time per iteration (s): 0.30 | learning rate: 3.631E-05 | global batch size: 256 | lm loss: 3.354834E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.935 | TFLOPs: 30.31 | 7: iteration 17410/ 21553 | consumed samples: 4456960 | consumed tokens: 9127854080 | elapsed time per iteration (s): 0.30 | learning rate: 3.623E-05 | global batch size: 256 | lm loss: 3.356662E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.593 | TFLOPs: 30.34 | 7: iteration 17420/ 21553 | consumed samples: 4459520 | consumed tokens: 9133096960 | elapsed time per iteration (s): 0.30 | learning rate: 3.616E-05 | global batch size: 256 | lm loss: 3.353476E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.317 | TFLOPs: 30.33 | 7: iteration 17430/ 21553 | consumed samples: 4462080 | consumed tokens: 9138339840 | elapsed time per iteration (s): 0.30 | learning rate: 3.608E-05 | global batch size: 256 | lm loss: 3.349081E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.506 | TFLOPs: 30.33 | 7: iteration 17440/ 21553 | consumed samples: 4464640 | consumed tokens: 9143582720 | elapsed time per iteration (s): 0.30 | learning rate: 3.600E-05 | global batch size: 256 | lm loss: 3.349656E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.781 | TFLOPs: 30.34 | 7: iteration 17450/ 21553 | consumed samples: 4467200 | consumed tokens: 9148825600 | elapsed time per iteration (s): 0.30 | learning rate: 3.593E-05 | global batch size: 256 | lm loss: 3.357245E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.063 | TFLOPs: 30.32 | 7: iteration 17460/ 21553 | consumed samples: 4469760 | consumed tokens: 9154068480 | elapsed time per iteration (s): 0.30 | learning rate: 3.585E-05 | global batch size: 256 | lm loss: 3.349395E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.628 | TFLOPs: 30.34 | 7: iteration 17470/ 21553 | consumed samples: 4472320 | consumed tokens: 9159311360 | elapsed time per iteration (s): 0.30 | learning rate: 3.578E-05 | global batch size: 256 | lm loss: 3.361776E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.496 | TFLOPs: 30.33 | 7: iteration 17480/ 21553 | consumed samples: 4474880 | consumed tokens: 9164554240 | elapsed time per iteration (s): 0.30 | learning rate: 3.570E-05 | global batch size: 256 | lm loss: 3.355483E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.500 | TFLOPs: 30.33 | 7: iteration 17490/ 21553 | consumed samples: 4477440 | consumed tokens: 9169797120 | elapsed time per iteration (s): 0.30 | learning rate: 3.563E-05 | global batch size: 256 | lm loss: 3.350985E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.260 | TFLOPs: 30.33 | 7: iteration 17500/ 21553 | consumed samples: 4480000 | consumed tokens: 9175040000 | elapsed time per iteration (s): 0.30 | learning rate: 3.555E-05 | global batch size: 256 | lm loss: 3.350451E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.199 | TFLOPs: 30.32 | 7: iteration 17510/ 21553 | consumed samples: 4482560 | consumed tokens: 9180282880 | elapsed time per iteration (s): 0.30 | learning rate: 3.548E-05 | global batch size: 256 | lm loss: 3.355861E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.852 | TFLOPs: 30.31 | 7: iteration 17520/ 21553 | consumed samples: 4485120 | consumed tokens: 9185525760 | elapsed time per iteration (s): 0.30 | learning rate: 3.541E-05 | global batch size: 256 | lm loss: 3.357979E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.907 | TFLOPs: 30.35 | 7: iteration 17530/ 21553 | consumed samples: 4487680 | consumed tokens: 9190768640 | elapsed time per iteration (s): 0.30 | learning rate: 3.533E-05 | global batch size: 256 | lm loss: 3.351082E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.062 | TFLOPs: 30.32 | 7: iteration 17540/ 21553 | consumed samples: 4490240 | consumed tokens: 9196011520 | elapsed time per iteration (s): 0.30 | learning rate: 3.526E-05 | global batch size: 256 | lm loss: 3.343165E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.172 | TFLOPs: 30.32 | 7: iteration 17550/ 21553 | consumed samples: 4492800 | consumed tokens: 9201254400 | elapsed time per iteration (s): 0.30 | learning rate: 3.518E-05 | global batch size: 256 | lm loss: 3.351582E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.331 | TFLOPs: 30.33 | 7: iteration 17560/ 21553 | consumed samples: 4495360 | consumed tokens: 9206497280 | elapsed time per iteration (s): 0.30 | learning rate: 3.511E-05 | global batch size: 256 | lm loss: 3.349463E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.136 | TFLOPs: 30.32 | 7: iteration 17570/ 21553 | consumed samples: 4497920 | consumed tokens: 9211740160 | elapsed time per iteration (s): 0.30 | learning rate: 3.504E-05 | global batch size: 256 | lm loss: 3.351177E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.236 | TFLOPs: 30.32 | 7: iteration 17580/ 21553 | consumed samples: 4500480 | consumed tokens: 9216983040 | elapsed time per iteration (s): 0.30 | learning rate: 3.496E-05 | global batch size: 256 | lm loss: 3.355187E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.386 | TFLOPs: 30.33 | 7: iteration 17590/ 21553 | consumed samples: 4503040 | consumed tokens: 9222225920 | elapsed time per iteration (s): 0.30 | learning rate: 3.489E-05 | global batch size: 256 | lm loss: 3.344614E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.186 | TFLOPs: 30.32 | 7: iteration 17600/ 21553 | consumed samples: 4505600 | consumed tokens: 9227468800 | elapsed time per iteration (s): 0.30 | learning rate: 3.482E-05 | global batch size: 256 | lm loss: 3.349900E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.300 | TFLOPs: 30.33 | 7: iteration 17610/ 21553 | consumed samples: 4508160 | consumed tokens: 9232711680 | elapsed time per iteration (s): 0.30 | learning rate: 3.475E-05 | global batch size: 256 | lm loss: 3.349829E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.732 | TFLOPs: 30.31 | 7: iteration 17620/ 21553 | consumed samples: 4510720 | consumed tokens: 9237954560 | elapsed time per iteration (s): 0.30 | learning rate: 3.467E-05 | global batch size: 256 | lm loss: 3.345747E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.721 | TFLOPs: 30.31 | 7: iteration 17630/ 21553 | consumed samples: 4513280 | consumed tokens: 9243197440 | elapsed time per iteration (s): 0.30 | learning rate: 3.460E-05 | global batch size: 256 | lm loss: 3.348701E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.491 | TFLOPs: 30.33 | 7: iteration 17640/ 21553 | consumed samples: 4515840 | consumed tokens: 9248440320 | elapsed time per iteration (s): 0.30 | learning rate: 3.453E-05 | global batch size: 256 | lm loss: 3.351498E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.989 | TFLOPs: 30.32 | 7: iteration 17650/ 21553 | consumed samples: 4518400 | consumed tokens: 9253683200 | elapsed time per iteration (s): 0.30 | learning rate: 3.446E-05 | global batch size: 256 | lm loss: 3.351593E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.023 | TFLOPs: 30.32 | 7: iteration 17660/ 21553 | consumed samples: 4520960 | consumed tokens: 9258926080 | elapsed time per iteration (s): 0.30 | learning rate: 3.438E-05 | global batch size: 256 | lm loss: 3.348586E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.013 | TFLOPs: 30.32 | 7: iteration 17670/ 21553 | consumed samples: 4523520 | consumed tokens: 9264168960 | elapsed time per iteration (s): 0.30 | learning rate: 3.431E-05 | global batch size: 256 | lm loss: 3.349081E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.211 | TFLOPs: 30.32 | 7: iteration 17680/ 21553 | consumed samples: 4526080 | consumed tokens: 9269411840 | elapsed time per iteration (s): 0.30 | learning rate: 3.424E-05 | global batch size: 256 | lm loss: 3.346386E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.790 | TFLOPs: 30.34 | 7: iteration 17690/ 21553 | consumed samples: 4528640 | consumed tokens: 9274654720 | elapsed time per iteration (s): 0.30 | learning rate: 3.417E-05 | global batch size: 256 | lm loss: 3.353677E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.585 | TFLOPs: 30.34 | 7: iteration 17700/ 21553 | consumed samples: 4531200 | consumed tokens: 9279897600 | elapsed time per iteration (s): 0.30 | learning rate: 3.410E-05 | global batch size: 256 | lm loss: 3.354863E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.751 | TFLOPs: 30.34 | 7: iteration 17710/ 21553 | consumed samples: 4533760 | consumed tokens: 9285140480 | elapsed time per iteration (s): 0.30 | learning rate: 3.403E-05 | global batch size: 256 | lm loss: 3.352946E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.907 | TFLOPs: 30.35 | 7: iteration 17720/ 21553 | consumed samples: 4536320 | consumed tokens: 9290383360 | elapsed time per iteration (s): 0.30 | learning rate: 3.396E-05 | global batch size: 256 | lm loss: 3.351216E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.429 | TFLOPs: 30.33 | 7: iteration 17730/ 21553 | consumed samples: 4538880 | consumed tokens: 9295626240 | elapsed time per iteration (s): 0.30 | learning rate: 3.388E-05 | global batch size: 256 | lm loss: 3.350248E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.178 | TFLOPs: 30.32 | 7: iteration 17740/ 21553 | consumed samples: 4541440 | consumed tokens: 9300869120 | elapsed time per iteration (s): 0.30 | learning rate: 3.381E-05 | global batch size: 256 | lm loss: 3.349254E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.892 | TFLOPs: 30.31 | 7: iteration 17750/ 21553 | consumed samples: 4544000 | consumed tokens: 9306112000 | elapsed time per iteration (s): 0.30 | learning rate: 3.374E-05 | global batch size: 256 | lm loss: 3.353792E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.144 | TFLOPs: 30.32 | 7: iteration 17760/ 21553 | consumed samples: 4546560 | consumed tokens: 9311354880 | elapsed time per iteration (s): 0.30 | learning rate: 3.367E-05 | global batch size: 256 | lm loss: 3.345814E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.910 | TFLOPs: 30.31 | 7: iteration 17770/ 21553 | consumed samples: 4549120 | consumed tokens: 9316597760 | elapsed time per iteration (s): 0.30 | learning rate: 3.360E-05 | global batch size: 256 | lm loss: 3.344270E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.650 | TFLOPs: 30.30 | 7: iteration 17780/ 21553 | consumed samples: 4551680 | consumed tokens: 9321840640 | elapsed time per iteration (s): 0.30 | learning rate: 3.353E-05 | global batch size: 256 | lm loss: 3.353581E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.039 | TFLOPs: 30.32 | 7: iteration 17790/ 21553 | consumed samples: 4554240 | consumed tokens: 9327083520 | elapsed time per iteration (s): 0.30 | learning rate: 3.346E-05 | global batch size: 256 | lm loss: 3.349477E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.122 | TFLOPs: 30.32 | 7: iteration 17800/ 21553 | consumed samples: 4556800 | consumed tokens: 9332326400 | elapsed time per iteration (s): 0.30 | learning rate: 3.339E-05 | global batch size: 256 | lm loss: 3.358192E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.826 | TFLOPs: 30.31 | 7: iteration 17810/ 21553 | consumed samples: 4559360 | consumed tokens: 9337569280 | elapsed time per iteration (s): 0.30 | learning rate: 3.332E-05 | global batch size: 256 | lm loss: 3.345422E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.759 | TFLOPs: 30.31 | 7: iteration 17820/ 21553 | consumed samples: 4561920 | consumed tokens: 9342812160 | elapsed time per iteration (s): 0.30 | learning rate: 3.326E-05 | global batch size: 256 | lm loss: 3.348745E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.751 | TFLOPs: 30.31 | 7: iteration 17830/ 21553 | consumed samples: 4564480 | consumed tokens: 9348055040 | elapsed time per iteration (s): 0.30 | learning rate: 3.319E-05 | global batch size: 256 | lm loss: 3.351141E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.264 | TFLOPs: 30.33 | 7: iteration 17840/ 21553 | consumed samples: 4567040 | consumed tokens: 9353297920 | elapsed time per iteration (s): 0.30 | learning rate: 3.312E-05 | global batch size: 256 | lm loss: 3.349450E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.846 | TFLOPs: 30.31 | 7: iteration 17850/ 21553 | consumed samples: 4569600 | consumed tokens: 9358540800 | elapsed time per iteration (s): 0.30 | learning rate: 3.305E-05 | global batch size: 256 | lm loss: 3.352393E+00 | grad norm: 0.446 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.336 | TFLOPs: 30.33 | 7: iteration 17860/ 21553 | consumed samples: 4572160 | consumed tokens: 9363783680 | elapsed time per iteration (s): 0.30 | learning rate: 3.298E-05 | global batch size: 256 | lm loss: 3.350924E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.913 | TFLOPs: 30.31 | 7: iteration 17870/ 21553 | consumed samples: 4574720 | consumed tokens: 9369026560 | elapsed time per iteration (s): 0.30 | learning rate: 3.291E-05 | global batch size: 256 | lm loss: 3.344358E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.149 | TFLOPs: 30.32 | 7: iteration 17880/ 21553 | consumed samples: 4577280 | consumed tokens: 9374269440 | elapsed time per iteration (s): 0.30 | learning rate: 3.284E-05 | global batch size: 256 | lm loss: 3.348587E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.428 | TFLOPs: 30.33 | 7: iteration 17890/ 21553 | consumed samples: 4579840 | consumed tokens: 9379512320 | elapsed time per iteration (s): 0.30 | learning rate: 3.277E-05 | global batch size: 256 | lm loss: 3.353328E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.830 | TFLOPs: 30.35 | 7: iteration 17900/ 21553 | consumed samples: 4582400 | consumed tokens: 9384755200 | elapsed time per iteration (s): 0.30 | learning rate: 3.271E-05 | global batch size: 256 | lm loss: 3.345189E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.184 | TFLOPs: 30.32 | 7: iteration 17910/ 21553 | consumed samples: 4584960 | consumed tokens: 9389998080 | elapsed time per iteration (s): 0.30 | learning rate: 3.264E-05 | global batch size: 256 | lm loss: 3.351323E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.069 | TFLOPs: 30.32 | 7: iteration 17920/ 21553 | consumed samples: 4587520 | consumed tokens: 9395240960 | elapsed time per iteration (s): 0.30 | learning rate: 3.257E-05 | global batch size: 256 | lm loss: 3.346358E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.120 | TFLOPs: 30.32 | 7: iteration 17930/ 21553 | consumed samples: 4590080 | consumed tokens: 9400483840 | elapsed time per iteration (s): 0.30 | learning rate: 3.250E-05 | global batch size: 256 | lm loss: 3.345192E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.139 | TFLOPs: 30.32 | 7: iteration 17940/ 21553 | consumed samples: 4592640 | consumed tokens: 9405726720 | elapsed time per iteration (s): 0.30 | learning rate: 3.244E-05 | global batch size: 256 | lm loss: 3.349950E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.260 | TFLOPs: 30.33 | 7: iteration 17950/ 21553 | consumed samples: 4595200 | consumed tokens: 9410969600 | elapsed time per iteration (s): 0.30 | learning rate: 3.237E-05 | global batch size: 256 | lm loss: 3.347966E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.220 | TFLOPs: 30.32 | 7: iteration 17960/ 21553 | consumed samples: 4597760 | consumed tokens: 9416212480 | elapsed time per iteration (s): 0.30 | learning rate: 3.230E-05 | global batch size: 256 | lm loss: 3.343227E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.248 | TFLOPs: 30.32 | 7: iteration 17970/ 21553 | consumed samples: 4600320 | consumed tokens: 9421455360 | elapsed time per iteration (s): 0.30 | learning rate: 3.224E-05 | global batch size: 256 | lm loss: 3.350956E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.640 | TFLOPs: 30.34 | 7: iteration 17980/ 21553 | consumed samples: 4602880 | consumed tokens: 9426698240 | elapsed time per iteration (s): 0.30 | learning rate: 3.217E-05 | global batch size: 256 | lm loss: 3.346027E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.715 | TFLOPs: 30.34 | 7: iteration 17990/ 21553 | consumed samples: 4605440 | consumed tokens: 9431941120 | elapsed time per iteration (s): 0.30 | learning rate: 3.210E-05 | global batch size: 256 | lm loss: 3.345130E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.351 | TFLOPs: 30.33 | 0: [2023-03-17 01:25:09,839] [INFO] [logging.py:68:log_dist] [Rank 0] step=18000, skipped=0, lr=[3.2036439682204886e-05, 3.2036439682204886e-05, 3.2036439682204886e-05], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 18000/ 21553 | consumed samples: 4608000 | consumed tokens: 9437184000 | elapsed time per iteration (s): 0.30 | learning rate: 3.204E-05 | global batch size: 256 | lm loss: 3.344736E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.680 | TFLOPs: 30.34 | 0: steps: 18000 loss: 3.3391 iter time (s): 0.294 samples/sec: 870.590 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 18000 | lm loss value: 3.928077E+00 | lm loss PPL: 5.080915E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 18000 to checkpoints_146m14b100mdedup 0: [2023-03-17 01:25:09,960] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step18000 is begin to save! 0: [2023-03-17 01:25:09,963] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:25:10,048] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:25:10,049] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:25:10,065] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:25:10,065] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:25:10,079] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:25:10,080] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:25:10,094] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:25:10,095] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:25:10,109] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:25:10,109] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:25:10,124] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:25:10,124] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:25:10,139] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:25:10,139] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:25:10,154] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:25:10,154] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:25:10,168] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:25:10,169] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:25:10,183] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:25:10,183] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:25:10,198] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:25:10,198] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:25:10,213] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:25:10,213] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:25:10,228] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:25:10,228] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:25:10,242] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:25:10,243] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:25:10,257] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:25:10,257] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:25:10,272] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:25:10,272] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:25:10,273] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:25:10,274] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step18000/mp_rank_00_model_states.pt 0: [2023-03-17 01:25:10,274] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/mp_rank_00_model_states.pt... 0: [2023-03-17 01:25:10,276] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/mp_rank_00_model_states.pt. 0: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:25:10,293] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:25:10,327] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:25:10,327] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 1: [2023-03-17 01:25:10,327] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:25:10,327] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 1: [2023-03-17 01:25:10,327] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 01:25:10,327] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 2: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:25:10,328] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 01:25:10,328] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 5: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 3: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:25:10,328] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 3: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:25:10,329] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 1: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:25:10,329] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 01:25:10,330] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,328] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:25:10,329] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 3: [2023-03-17 01:25:10,330] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:25:10,331] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 3: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:25:10,331] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 0: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:25:10,331] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:25:10,331] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:25:10,332] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,332] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 1: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:25:10,332] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 3: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:25:10,332] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 01:25:10,332] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 3: [2023-03-17 01:25:10,332] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:25:10,333] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 1: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:25:10,333] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 1: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:25:10,333] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 5: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 1: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 5: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 1: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 5: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 5: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 1: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,328] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 7: [2023-03-17 01:25:10,331] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 4: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 7: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 4: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,329] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 7: [2023-03-17 01:25:10,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:25:10,333] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 4: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 7: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 7: [2023-03-17 01:25:10,333] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 4: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 4: [2023-03-17 01:25:10,331] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 7: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,331] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 7: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 7: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 1: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 4: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 4: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 4: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 0: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:25:10,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 0: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:25:10,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 0: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:25:10,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 0: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:25:10,336] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 3: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:25:10,336] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 7: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:25:10,335] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,334] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:25:10,334] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,335] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:25:10,336] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 6: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:25:10,336] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 5: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 5: [2023-03-17 01:25:10,336] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 2: [2023-03-17 01:25:10,328] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 01:25:10,328] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 2: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 2: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:25:10,328] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 01:25:10,328] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 2: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:25:10,329] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:25:10,329] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 2: [2023-03-17 01:25:10,330] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:25:10,330] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 01:25:10,330] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 2: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:25:10,336] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 2: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:25:10,336] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 2: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:25:10,336] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 0: [2023-03-17 01:25:10,337] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:25:10,337] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 01:25:10,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 4: [2023-03-17 01:25:10,336] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 3: [2023-03-17 01:25:10,338] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,336] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 3: [2023-03-17 01:25:10,338] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 01:25:10,338] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 7: [2023-03-17 01:25:10,339] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:25:10,339] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 01:25:10,339] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 4: [2023-03-17 01:25:10,340] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:25:10,340] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:25:10,340] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 5: [2023-03-17 01:25:10,341] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:25:10,342] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 01:25:10,342] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 1: [2023-03-17 01:25:10,349] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:25:10,349] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 01:25:10,349] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 0: [2023-03-17 01:25:10,357] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step18000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:25:10,357] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step18000 is ready now! 0: successfully saved checkpoint at iteration 18000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 402.47 7: iteration 18010/ 21553 | consumed samples: 4610560 | consumed tokens: 9442426880 | elapsed time per iteration (s): 0.35 | learning rate: 3.197E-05 | global batch size: 256 | lm loss: 3.345460E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 738.245 | TFLOPs: 25.84 | 7: iteration 18020/ 21553 | consumed samples: 4613120 | consumed tokens: 9447669760 | elapsed time per iteration (s): 0.30 | learning rate: 3.190E-05 | global batch size: 256 | lm loss: 3.339973E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.476 | TFLOPs: 30.33 | 7: iteration 18030/ 21553 | consumed samples: 4615680 | consumed tokens: 9452912640 | elapsed time per iteration (s): 0.30 | learning rate: 3.184E-05 | global batch size: 256 | lm loss: 3.345681E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.011 | TFLOPs: 30.32 | 7: iteration 18040/ 21553 | consumed samples: 4618240 | consumed tokens: 9458155520 | elapsed time per iteration (s): 0.30 | learning rate: 3.177E-05 | global batch size: 256 | lm loss: 3.348998E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.555 | TFLOPs: 30.34 | 7: iteration 18050/ 21553 | consumed samples: 4620800 | consumed tokens: 9463398400 | elapsed time per iteration (s): 0.30 | learning rate: 3.171E-05 | global batch size: 256 | lm loss: 3.339539E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.862 | TFLOPs: 30.35 | 7: iteration 18060/ 21553 | consumed samples: 4623360 | consumed tokens: 9468641280 | elapsed time per iteration (s): 0.30 | learning rate: 3.164E-05 | global batch size: 256 | lm loss: 3.342535E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.560 | TFLOPs: 30.34 | 7: iteration 18070/ 21553 | consumed samples: 4625920 | consumed tokens: 9473884160 | elapsed time per iteration (s): 0.30 | learning rate: 3.158E-05 | global batch size: 256 | lm loss: 3.346083E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.476 | TFLOPs: 30.33 | 7: iteration 18080/ 21553 | consumed samples: 4628480 | consumed tokens: 9479127040 | elapsed time per iteration (s): 0.30 | learning rate: 3.151E-05 | global batch size: 256 | lm loss: 3.348215E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.394 | TFLOPs: 30.37 | 7: iteration 18090/ 21553 | consumed samples: 4631040 | consumed tokens: 9484369920 | elapsed time per iteration (s): 0.30 | learning rate: 3.145E-05 | global batch size: 256 | lm loss: 3.347267E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.307 | TFLOPs: 30.36 | 7: iteration 18100/ 21553 | consumed samples: 4633600 | consumed tokens: 9489612800 | elapsed time per iteration (s): 0.30 | learning rate: 3.138E-05 | global batch size: 256 | lm loss: 3.347012E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.723 | TFLOPs: 30.38 | 7: iteration 18110/ 21553 | consumed samples: 4636160 | consumed tokens: 9494855680 | elapsed time per iteration (s): 0.30 | learning rate: 3.132E-05 | global batch size: 256 | lm loss: 3.344239E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 862.703 | TFLOPs: 30.20 | 7: iteration 18120/ 21553 | consumed samples: 4638720 | consumed tokens: 9500098560 | elapsed time per iteration (s): 0.30 | learning rate: 3.125E-05 | global batch size: 256 | lm loss: 3.349959E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.076 | TFLOPs: 30.35 | 7: iteration 18130/ 21553 | consumed samples: 4641280 | consumed tokens: 9505341440 | elapsed time per iteration (s): 0.30 | learning rate: 3.119E-05 | global batch size: 256 | lm loss: 3.341458E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.513 | TFLOPs: 30.33 | 7: iteration 18140/ 21553 | consumed samples: 4643840 | consumed tokens: 9510584320 | elapsed time per iteration (s): 0.30 | learning rate: 3.113E-05 | global batch size: 256 | lm loss: 3.350080E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.612 | TFLOPs: 30.34 | 7: iteration 18150/ 21553 | consumed samples: 4646400 | consumed tokens: 9515827200 | elapsed time per iteration (s): 0.30 | learning rate: 3.106E-05 | global batch size: 256 | lm loss: 3.339814E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.799 | TFLOPs: 30.34 | 7: iteration 18160/ 21553 | consumed samples: 4648960 | consumed tokens: 9521070080 | elapsed time per iteration (s): 0.30 | learning rate: 3.100E-05 | global batch size: 256 | lm loss: 3.345383E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.426 | TFLOPs: 30.33 | 7: iteration 18170/ 21553 | consumed samples: 4651520 | consumed tokens: 9526312960 | elapsed time per iteration (s): 0.30 | learning rate: 3.094E-05 | global batch size: 256 | lm loss: 3.342894E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.274 | TFLOPs: 30.33 | 7: iteration 18180/ 21553 | consumed samples: 4654080 | consumed tokens: 9531555840 | elapsed time per iteration (s): 0.30 | learning rate: 3.087E-05 | global batch size: 256 | lm loss: 3.345925E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.680 | TFLOPs: 30.34 | 7: iteration 18190/ 21553 | consumed samples: 4656640 | consumed tokens: 9536798720 | elapsed time per iteration (s): 0.30 | learning rate: 3.081E-05 | global batch size: 256 | lm loss: 3.344708E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.114 | TFLOPs: 30.36 | 7: iteration 18200/ 21553 | consumed samples: 4659200 | consumed tokens: 9542041600 | elapsed time per iteration (s): 0.30 | learning rate: 3.075E-05 | global batch size: 256 | lm loss: 3.345811E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.193 | TFLOPs: 30.36 | 7: iteration 18210/ 21553 | consumed samples: 4661760 | consumed tokens: 9547284480 | elapsed time per iteration (s): 0.30 | learning rate: 3.068E-05 | global batch size: 256 | lm loss: 3.345610E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.697 | TFLOPs: 30.34 | 7: iteration 18220/ 21553 | consumed samples: 4664320 | consumed tokens: 9552527360 | elapsed time per iteration (s): 0.30 | learning rate: 3.062E-05 | global batch size: 256 | lm loss: 3.348460E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.283 | TFLOPs: 30.33 | 7: iteration 18230/ 21553 | consumed samples: 4666880 | consumed tokens: 9557770240 | elapsed time per iteration (s): 0.30 | learning rate: 3.056E-05 | global batch size: 256 | lm loss: 3.341301E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.932 | TFLOPs: 30.31 | 7: iteration 18240/ 21553 | consumed samples: 4669440 | consumed tokens: 9563013120 | elapsed time per iteration (s): 0.30 | learning rate: 3.050E-05 | global batch size: 256 | lm loss: 3.344642E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.625 | TFLOPs: 30.34 | 7: iteration 18250/ 21553 | consumed samples: 4672000 | consumed tokens: 9568256000 | elapsed time per iteration (s): 0.30 | learning rate: 3.043E-05 | global batch size: 256 | lm loss: 3.339553E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.696 | TFLOPs: 30.34 | 7: iteration 18260/ 21553 | consumed samples: 4674560 | consumed tokens: 9573498880 | elapsed time per iteration (s): 0.30 | learning rate: 3.037E-05 | global batch size: 256 | lm loss: 3.338457E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.590 | TFLOPs: 30.34 | 7: iteration 18270/ 21553 | consumed samples: 4677120 | consumed tokens: 9578741760 | elapsed time per iteration (s): 0.30 | learning rate: 3.031E-05 | global batch size: 256 | lm loss: 3.338793E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.900 | TFLOPs: 30.35 | 7: iteration 18280/ 21553 | consumed samples: 4679680 | consumed tokens: 9583984640 | elapsed time per iteration (s): 0.30 | learning rate: 3.025E-05 | global batch size: 256 | lm loss: 3.344366E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.719 | TFLOPs: 30.34 | 7: iteration 18290/ 21553 | consumed samples: 4682240 | consumed tokens: 9589227520 | elapsed time per iteration (s): 0.30 | learning rate: 3.019E-05 | global batch size: 256 | lm loss: 3.343679E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.449 | TFLOPs: 30.33 | 7: iteration 18300/ 21553 | consumed samples: 4684800 | consumed tokens: 9594470400 | elapsed time per iteration (s): 0.30 | learning rate: 3.013E-05 | global batch size: 256 | lm loss: 3.347749E+00 | grad norm: 0.441 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.154 | TFLOPs: 30.36 | 7: iteration 18310/ 21553 | consumed samples: 4687360 | consumed tokens: 9599713280 | elapsed time per iteration (s): 0.30 | learning rate: 3.007E-05 | global batch size: 256 | lm loss: 3.340342E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.997 | TFLOPs: 30.35 | 7: iteration 18320/ 21553 | consumed samples: 4689920 | consumed tokens: 9604956160 | elapsed time per iteration (s): 0.30 | learning rate: 3.001E-05 | global batch size: 256 | lm loss: 3.342133E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.811 | TFLOPs: 30.34 | 7: iteration 18330/ 21553 | consumed samples: 4692480 | consumed tokens: 9610199040 | elapsed time per iteration (s): 0.30 | learning rate: 2.994E-05 | global batch size: 256 | lm loss: 3.345719E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.710 | TFLOPs: 30.34 | 7: iteration 18340/ 21553 | consumed samples: 4695040 | consumed tokens: 9615441920 | elapsed time per iteration (s): 0.30 | learning rate: 2.988E-05 | global batch size: 256 | lm loss: 3.342780E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.243 | TFLOPs: 30.36 | 7: iteration 18350/ 21553 | consumed samples: 4697600 | consumed tokens: 9620684800 | elapsed time per iteration (s): 0.30 | learning rate: 2.982E-05 | global batch size: 256 | lm loss: 3.345987E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.380 | TFLOPs: 30.33 | 7: iteration 18360/ 21553 | consumed samples: 4700160 | consumed tokens: 9625927680 | elapsed time per iteration (s): 0.30 | learning rate: 2.976E-05 | global batch size: 256 | lm loss: 3.353201E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.214 | TFLOPs: 30.32 | 7: iteration 18370/ 21553 | consumed samples: 4702720 | consumed tokens: 9631170560 | elapsed time per iteration (s): 0.30 | learning rate: 2.970E-05 | global batch size: 256 | lm loss: 3.341804E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.508 | TFLOPs: 30.30 | 7: iteration 18380/ 21553 | consumed samples: 4705280 | consumed tokens: 9636413440 | elapsed time per iteration (s): 0.30 | learning rate: 2.964E-05 | global batch size: 256 | lm loss: 3.343432E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.757 | TFLOPs: 30.31 | 7: iteration 18390/ 21553 | consumed samples: 4707840 | consumed tokens: 9641656320 | elapsed time per iteration (s): 0.30 | learning rate: 2.958E-05 | global batch size: 256 | lm loss: 3.345572E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.102 | TFLOPs: 30.32 | 7: iteration 18400/ 21553 | consumed samples: 4710400 | consumed tokens: 9646899200 | elapsed time per iteration (s): 0.30 | learning rate: 2.953E-05 | global batch size: 256 | lm loss: 3.342729E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.659 | TFLOPs: 30.30 | 7: iteration 18410/ 21553 | consumed samples: 4712960 | consumed tokens: 9652142080 | elapsed time per iteration (s): 0.30 | learning rate: 2.947E-05 | global batch size: 256 | lm loss: 3.340199E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.895 | TFLOPs: 30.31 | 7: iteration 18420/ 21553 | consumed samples: 4715520 | consumed tokens: 9657384960 | elapsed time per iteration (s): 0.30 | learning rate: 2.941E-05 | global batch size: 256 | lm loss: 3.344325E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.809 | TFLOPs: 30.31 | 7: iteration 18430/ 21553 | consumed samples: 4718080 | consumed tokens: 9662627840 | elapsed time per iteration (s): 0.30 | learning rate: 2.935E-05 | global batch size: 256 | lm loss: 3.350828E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.860 | TFLOPs: 30.31 | 7: iteration 18440/ 21553 | consumed samples: 4720640 | consumed tokens: 9667870720 | elapsed time per iteration (s): 0.30 | learning rate: 2.929E-05 | global batch size: 256 | lm loss: 3.343654E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.358 | TFLOPs: 30.29 | 7: iteration 18450/ 21553 | consumed samples: 4723200 | consumed tokens: 9673113600 | elapsed time per iteration (s): 0.30 | learning rate: 2.923E-05 | global batch size: 256 | lm loss: 3.339532E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.817 | TFLOPs: 30.31 | 7: iteration 18460/ 21553 | consumed samples: 4725760 | consumed tokens: 9678356480 | elapsed time per iteration (s): 0.30 | learning rate: 2.917E-05 | global batch size: 256 | lm loss: 3.348587E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.828 | TFLOPs: 30.31 | 7: iteration 18470/ 21553 | consumed samples: 4728320 | consumed tokens: 9683599360 | elapsed time per iteration (s): 0.30 | learning rate: 2.911E-05 | global batch size: 256 | lm loss: 3.339127E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.520 | TFLOPs: 30.30 | 7: iteration 18480/ 21553 | consumed samples: 4730880 | consumed tokens: 9688842240 | elapsed time per iteration (s): 0.30 | learning rate: 2.906E-05 | global batch size: 256 | lm loss: 3.338955E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.645 | TFLOPs: 30.30 | 7: iteration 18490/ 21553 | consumed samples: 4733440 | consumed tokens: 9694085120 | elapsed time per iteration (s): 0.30 | learning rate: 2.900E-05 | global batch size: 256 | lm loss: 3.340744E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.564 | TFLOPs: 30.30 | 7: iteration 18500/ 21553 | consumed samples: 4736000 | consumed tokens: 9699328000 | elapsed time per iteration (s): 0.30 | learning rate: 2.894E-05 | global batch size: 256 | lm loss: 3.340137E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.723 | TFLOPs: 30.31 | 7: iteration 18510/ 21553 | consumed samples: 4738560 | consumed tokens: 9704570880 | elapsed time per iteration (s): 0.30 | learning rate: 2.888E-05 | global batch size: 256 | lm loss: 3.348595E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.805 | TFLOPs: 30.31 | 7: iteration 18520/ 21553 | consumed samples: 4741120 | consumed tokens: 9709813760 | elapsed time per iteration (s): 0.30 | learning rate: 2.883E-05 | global batch size: 256 | lm loss: 3.335944E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.081 | TFLOPs: 30.32 | 7: iteration 18530/ 21553 | consumed samples: 4743680 | consumed tokens: 9715056640 | elapsed time per iteration (s): 0.30 | learning rate: 2.877E-05 | global batch size: 256 | lm loss: 3.343345E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.079 | TFLOPs: 30.32 | 7: iteration 18540/ 21553 | consumed samples: 4746240 | consumed tokens: 9720299520 | elapsed time per iteration (s): 0.30 | learning rate: 2.871E-05 | global batch size: 256 | lm loss: 3.346916E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.060 | TFLOPs: 30.32 | 7: iteration 18550/ 21553 | consumed samples: 4748800 | consumed tokens: 9725542400 | elapsed time per iteration (s): 0.30 | learning rate: 2.865E-05 | global batch size: 256 | lm loss: 3.341906E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.242 | TFLOPs: 30.32 | 7: iteration 18560/ 21553 | consumed samples: 4751360 | consumed tokens: 9730785280 | elapsed time per iteration (s): 0.30 | learning rate: 2.860E-05 | global batch size: 256 | lm loss: 3.339064E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.795 | TFLOPs: 30.31 | 7: iteration 18570/ 21553 | consumed samples: 4753920 | consumed tokens: 9736028160 | elapsed time per iteration (s): 0.30 | learning rate: 2.854E-05 | global batch size: 256 | lm loss: 3.339686E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.808 | TFLOPs: 30.31 | 7: iteration 18580/ 21553 | consumed samples: 4756480 | consumed tokens: 9741271040 | elapsed time per iteration (s): 0.31 | learning rate: 2.849E-05 | global batch size: 256 | lm loss: 3.341869E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 825.436 | TFLOPs: 28.90 | 7: iteration 18590/ 21553 | consumed samples: 4759040 | consumed tokens: 9746513920 | elapsed time per iteration (s): 0.30 | learning rate: 2.843E-05 | global batch size: 256 | lm loss: 3.346719E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.381 | TFLOPs: 30.33 | 7: iteration 18600/ 21553 | consumed samples: 4761600 | consumed tokens: 9751756800 | elapsed time per iteration (s): 0.30 | learning rate: 2.837E-05 | global batch size: 256 | lm loss: 3.346611E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.449 | TFLOPs: 30.33 | 7: iteration 18610/ 21553 | consumed samples: 4764160 | consumed tokens: 9756999680 | elapsed time per iteration (s): 0.30 | learning rate: 2.832E-05 | global batch size: 256 | lm loss: 3.346752E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.547 | TFLOPs: 30.34 | 7: iteration 18620/ 21553 | consumed samples: 4766720 | consumed tokens: 9762242560 | elapsed time per iteration (s): 0.30 | learning rate: 2.826E-05 | global batch size: 256 | lm loss: 3.341114E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.188 | TFLOPs: 30.32 | 7: iteration 18630/ 21553 | consumed samples: 4769280 | consumed tokens: 9767485440 | elapsed time per iteration (s): 0.30 | learning rate: 2.821E-05 | global batch size: 256 | lm loss: 3.342906E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.365 | TFLOPs: 30.33 | 7: iteration 18640/ 21553 | consumed samples: 4771840 | consumed tokens: 9772728320 | elapsed time per iteration (s): 0.30 | learning rate: 2.815E-05 | global batch size: 256 | lm loss: 3.337285E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.509 | TFLOPs: 30.33 | 7: iteration 18650/ 21553 | consumed samples: 4774400 | consumed tokens: 9777971200 | elapsed time per iteration (s): 0.30 | learning rate: 2.810E-05 | global batch size: 256 | lm loss: 3.343806E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.059 | TFLOPs: 30.35 | 7: iteration 18660/ 21553 | consumed samples: 4776960 | consumed tokens: 9783214080 | elapsed time per iteration (s): 0.30 | learning rate: 2.804E-05 | global batch size: 256 | lm loss: 3.343754E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.735 | TFLOPs: 30.34 | 7: iteration 18670/ 21553 | consumed samples: 4779520 | consumed tokens: 9788456960 | elapsed time per iteration (s): 0.30 | learning rate: 2.799E-05 | global batch size: 256 | lm loss: 3.343812E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.352 | TFLOPs: 30.33 | 7: iteration 18680/ 21553 | consumed samples: 4782080 | consumed tokens: 9793699840 | elapsed time per iteration (s): 0.30 | learning rate: 2.793E-05 | global batch size: 256 | lm loss: 3.344262E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.622 | TFLOPs: 30.34 | 7: iteration 18690/ 21553 | consumed samples: 4784640 | consumed tokens: 9798942720 | elapsed time per iteration (s): 0.30 | learning rate: 2.788E-05 | global batch size: 256 | lm loss: 3.339797E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.022 | TFLOPs: 30.35 | 7: iteration 18700/ 21553 | consumed samples: 4787200 | consumed tokens: 9804185600 | elapsed time per iteration (s): 0.30 | learning rate: 2.782E-05 | global batch size: 256 | lm loss: 3.338110E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.135 | TFLOPs: 30.32 | 7: iteration 18710/ 21553 | consumed samples: 4789760 | consumed tokens: 9809428480 | elapsed time per iteration (s): 0.30 | learning rate: 2.777E-05 | global batch size: 256 | lm loss: 3.342457E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.220 | TFLOPs: 30.32 | 7: iteration 18720/ 21553 | consumed samples: 4792320 | consumed tokens: 9814671360 | elapsed time per iteration (s): 0.30 | learning rate: 2.772E-05 | global batch size: 256 | lm loss: 3.337060E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.561 | TFLOPs: 30.34 | 7: iteration 18730/ 21553 | consumed samples: 4794880 | consumed tokens: 9819914240 | elapsed time per iteration (s): 0.30 | learning rate: 2.766E-05 | global batch size: 256 | lm loss: 3.343437E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.743 | TFLOPs: 30.34 | 7: iteration 18740/ 21553 | consumed samples: 4797440 | consumed tokens: 9825157120 | elapsed time per iteration (s): 0.30 | learning rate: 2.761E-05 | global batch size: 256 | lm loss: 3.345019E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.750 | TFLOPs: 30.34 | 7: iteration 18750/ 21553 | consumed samples: 4800000 | consumed tokens: 9830400000 | elapsed time per iteration (s): 0.30 | learning rate: 2.756E-05 | global batch size: 256 | lm loss: 3.340135E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.857 | TFLOPs: 30.35 | 7: iteration 18760/ 21553 | consumed samples: 4802560 | consumed tokens: 9835642880 | elapsed time per iteration (s): 0.30 | learning rate: 2.750E-05 | global batch size: 256 | lm loss: 3.341865E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.915 | TFLOPs: 30.31 | 7: iteration 18770/ 21553 | consumed samples: 4805120 | consumed tokens: 9840885760 | elapsed time per iteration (s): 0.30 | learning rate: 2.745E-05 | global batch size: 256 | lm loss: 3.335680E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.831 | TFLOPs: 30.31 | 7: iteration 18780/ 21553 | consumed samples: 4807680 | consumed tokens: 9846128640 | elapsed time per iteration (s): 0.30 | learning rate: 2.740E-05 | global batch size: 256 | lm loss: 3.344894E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.794 | TFLOPs: 30.31 | 7: iteration 18790/ 21553 | consumed samples: 4810240 | consumed tokens: 9851371520 | elapsed time per iteration (s): 0.30 | learning rate: 2.735E-05 | global batch size: 256 | lm loss: 3.338256E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.059 | TFLOPs: 30.32 | 7: iteration 18800/ 21553 | consumed samples: 4812800 | consumed tokens: 9856614400 | elapsed time per iteration (s): 0.30 | learning rate: 2.729E-05 | global batch size: 256 | lm loss: 3.332357E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.773 | TFLOPs: 30.31 | 7: iteration 18810/ 21553 | consumed samples: 4815360 | consumed tokens: 9861857280 | elapsed time per iteration (s): 0.30 | learning rate: 2.724E-05 | global batch size: 256 | lm loss: 3.337574E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.377 | TFLOPs: 30.29 | 7: iteration 18820/ 21553 | consumed samples: 4817920 | consumed tokens: 9867100160 | elapsed time per iteration (s): 0.30 | learning rate: 2.719E-05 | global batch size: 256 | lm loss: 3.340539E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.581 | TFLOPs: 30.30 | 7: iteration 18830/ 21553 | consumed samples: 4820480 | consumed tokens: 9872343040 | elapsed time per iteration (s): 0.30 | learning rate: 2.714E-05 | global batch size: 256 | lm loss: 3.338153E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.080 | TFLOPs: 30.32 | 7: iteration 18840/ 21553 | consumed samples: 4823040 | consumed tokens: 9877585920 | elapsed time per iteration (s): 0.30 | learning rate: 2.709E-05 | global batch size: 256 | lm loss: 3.338520E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.914 | TFLOPs: 30.31 | 7: iteration 18850/ 21553 | consumed samples: 4825600 | consumed tokens: 9882828800 | elapsed time per iteration (s): 0.30 | learning rate: 2.703E-05 | global batch size: 256 | lm loss: 3.339188E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.243 | TFLOPs: 30.32 | 7: iteration 18860/ 21553 | consumed samples: 4828160 | consumed tokens: 9888071680 | elapsed time per iteration (s): 0.30 | learning rate: 2.698E-05 | global batch size: 256 | lm loss: 3.344310E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.092 | TFLOPs: 30.32 | 7: iteration 18870/ 21553 | consumed samples: 4830720 | consumed tokens: 9893314560 | elapsed time per iteration (s): 0.30 | learning rate: 2.693E-05 | global batch size: 256 | lm loss: 3.336070E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.752 | TFLOPs: 30.31 | 7: iteration 18880/ 21553 | consumed samples: 4833280 | consumed tokens: 9898557440 | elapsed time per iteration (s): 0.30 | learning rate: 2.688E-05 | global batch size: 256 | lm loss: 3.336514E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.086 | TFLOPs: 30.32 | 7: iteration 18890/ 21553 | consumed samples: 4835840 | consumed tokens: 9903800320 | elapsed time per iteration (s): 0.30 | learning rate: 2.683E-05 | global batch size: 256 | lm loss: 3.335521E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.932 | TFLOPs: 30.31 | 7: iteration 18900/ 21553 | consumed samples: 4838400 | consumed tokens: 9909043200 | elapsed time per iteration (s): 0.30 | learning rate: 2.678E-05 | global batch size: 256 | lm loss: 3.344770E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.453 | TFLOPs: 30.33 | 7: iteration 18910/ 21553 | consumed samples: 4840960 | consumed tokens: 9914286080 | elapsed time per iteration (s): 0.30 | learning rate: 2.673E-05 | global batch size: 256 | lm loss: 3.337024E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.717 | TFLOPs: 30.34 | 7: iteration 18920/ 21553 | consumed samples: 4843520 | consumed tokens: 9919528960 | elapsed time per iteration (s): 0.30 | learning rate: 2.668E-05 | global batch size: 256 | lm loss: 3.338613E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.817 | TFLOPs: 30.34 | 7: iteration 18930/ 21553 | consumed samples: 4846080 | consumed tokens: 9924771840 | elapsed time per iteration (s): 0.30 | learning rate: 2.663E-05 | global batch size: 256 | lm loss: 3.334292E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.493 | TFLOPs: 30.33 | 7: iteration 18940/ 21553 | consumed samples: 4848640 | consumed tokens: 9930014720 | elapsed time per iteration (s): 0.30 | learning rate: 2.658E-05 | global batch size: 256 | lm loss: 3.340070E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.271 | TFLOPs: 30.33 | 7: iteration 18950/ 21553 | consumed samples: 4851200 | consumed tokens: 9935257600 | elapsed time per iteration (s): 0.30 | learning rate: 2.653E-05 | global batch size: 256 | lm loss: 3.335614E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.656 | TFLOPs: 30.34 | 7: iteration 18960/ 21553 | consumed samples: 4853760 | consumed tokens: 9940500480 | elapsed time per iteration (s): 0.30 | learning rate: 2.648E-05 | global batch size: 256 | lm loss: 3.344464E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.797 | TFLOPs: 30.34 | 7: iteration 18970/ 21553 | consumed samples: 4856320 | consumed tokens: 9945743360 | elapsed time per iteration (s): 0.30 | learning rate: 2.643E-05 | global batch size: 256 | lm loss: 3.336827E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.155 | TFLOPs: 30.32 | 7: iteration 18980/ 21553 | consumed samples: 4858880 | consumed tokens: 9950986240 | elapsed time per iteration (s): 0.30 | learning rate: 2.638E-05 | global batch size: 256 | lm loss: 3.337368E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.311 | TFLOPs: 30.33 | 7: iteration 18990/ 21553 | consumed samples: 4861440 | consumed tokens: 9956229120 | elapsed time per iteration (s): 0.30 | learning rate: 2.633E-05 | global batch size: 256 | lm loss: 3.336154E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.346 | TFLOPs: 30.33 | 7: iteration 19000/ 21553 | consumed samples: 4864000 | consumed tokens: 9961472000 | elapsed time per iteration (s): 0.30 | learning rate: 2.628E-05 | global batch size: 256 | lm loss: 3.334565E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.390 | TFLOPs: 30.33 | 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 19000 | lm loss value: 3.934821E+00 | lm loss PPL: 5.115299E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 19000 to checkpoints_146m14b100mdedup 0: [2023-03-17 01:30:06,109] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step19000 is begin to save! 0: [2023-03-17 01:30:06,112] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:30:06,198] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:30:06,199] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:30:06,214] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:30:06,214] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:30:06,229] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:30:06,229] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:30:06,244] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:30:06,244] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:30:06,259] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:30:06,259] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:30:06,274] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:30:06,274] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:30:06,288] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:30:06,289] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:30:06,303] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:30:06,303] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:30:06,318] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:30:06,318] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:30:06,333] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:30:06,333] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:30:06,348] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:30:06,348] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:30:06,363] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:30:06,363] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:30:06,378] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:30:06,378] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:30:06,393] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:30:06,393] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:30:06,407] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:30:06,408] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:30:06,422] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:30:06,423] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:30:06,424] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:30:06,424] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step19000/mp_rank_00_model_states.pt 0: [2023-03-17 01:30:06,424] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/mp_rank_00_model_states.pt... 0: [2023-03-17 01:30:06,426] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/mp_rank_00_model_states.pt. 7: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:30:06,443] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:30:06,477] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:30:06,477] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:30:06,477] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 3: [2023-03-17 01:30:06,477] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 01:30:06,477] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 4: [2023-03-17 01:30:06,477] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 3: [2023-03-17 01:30:06,478] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:30:06,478] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 01:30:06,478] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 4: [2023-03-17 01:30:06,478] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:30:06,478] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:30:06,478] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 2: [2023-03-17 01:30:06,478] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:30:06,478] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:30:06,478] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 2: [2023-03-17 01:30:06,478] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 4: [2023-03-17 01:30:06,478] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 2: [2023-03-17 01:30:06,478] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: [2023-03-17 01:30:06,478] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:30:06,479] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 01:30:06,479] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 6: [2023-03-17 01:30:06,479] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:30:06,479] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:30:06,479] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 1: [2023-03-17 01:30:06,479] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:30:06,479] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,479] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 6: [2023-03-17 01:30:06,480] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:30:06,480] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 5: [2023-03-17 01:30:06,480] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:30:06,480] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 5: [2023-03-17 01:30:06,480] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 2: [2023-03-17 01:30:06,480] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:30:06,480] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 4: [2023-03-17 01:30:06,481] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:30:06,481] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,481] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:30:06,481] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 1: [2023-03-17 01:30:06,481] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,481] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 5: [2023-03-17 01:30:06,481] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:30:06,481] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 01:30:06,481] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 5: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:30:06,482] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 6: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:30:06,482] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 5: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:30:06,482] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 3: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:30:06,482] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 01:30:06,482] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:30:06,482] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 3: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: [2023-03-17 01:30:06,482] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 7: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:30:06,482] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 7: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:30:06,483] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 5: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:30:06,483] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 1: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:30:06,483] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,483] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 1: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 1: [2023-03-17 01:30:06,483] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 4: [2023-03-17 01:30:06,484] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:30:06,484] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 01:30:06,484] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 6: [2023-03-17 01:30:06,484] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:30:06,484] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 01:30:06,484] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 4: [2023-03-17 01:30:06,485] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:30:06,485] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:30:06,485] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 6: [2023-03-17 01:30:06,485] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:30:06,485] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 01:30:06,485] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 3: [2023-03-17 01:30:06,485] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:30:06,485] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 6: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 3: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 6: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 6: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 7: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 6: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 7: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 4: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 3: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 3: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 5: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:30:06,487] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 01:30:06,487] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 5: [2023-03-17 01:30:06,487] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:30:06,487] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 4: [2023-03-17 01:30:06,487] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:30:06,487] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 4: [2023-03-17 01:30:06,487] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 01:30:06,487] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:30:06,487] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 01:30:06,487] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 4: [2023-03-17 01:30:06,487] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 6: [2023-03-17 01:30:06,489] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:30:06,489] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:30:06,489] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 1: [2023-03-17 01:30:06,489] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:30:06,489] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,489] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 2: [2023-03-17 01:30:06,480] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,489] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:30:06,480] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 2: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:30:06,489] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 2: [2023-03-17 01:30:06,482] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 01:30:06,482] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 2: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:30:06,483] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,489] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 2: [2023-03-17 01:30:06,483] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 2: [2023-03-17 01:30:06,485] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:30:06,485] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 01:30:06,485] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 2: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 1: [2023-03-17 01:30:06,490] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:30:06,490] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 01:30:06,490] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 2: [2023-03-17 01:30:06,490] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:30:06,490] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:30:06,490] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 7: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 01:30:06,486] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 7: [2023-03-17 01:30:06,486] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 7: [2023-03-17 01:30:06,487] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:30:06,487] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:30:06,487] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 7: [2023-03-17 01:30:06,489] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:30:06,489] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 01:30:06,489] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 7: [2023-03-17 01:30:06,490] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:30:06,490] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 01:30:06,490] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: [2023-03-17 01:30:06,492] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:30:06,492] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:30:06,492] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 01:30:06,492] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:30:06,492] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 01:30:06,492] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: [2023-03-17 01:30:06,492] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: [2023-03-17 01:30:06,492] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 01:30:06,492] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 5: [2023-03-17 01:30:06,492] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:30:06,493] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 01:30:06,493] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: [2023-03-17 01:30:06,505] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:30:06,505] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:30:06,505] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:30:06,505] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 01:30:06,505] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: [2023-03-17 01:30:06,505] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: [2023-03-17 01:30:06,506] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step19000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:30:06,506] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step19000 is ready now! 0: successfully saved checkpoint at iteration 19000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 401.85 7: iteration 19010/ 21553 | consumed samples: 4866560 | consumed tokens: 9966714880 | elapsed time per iteration (s): 0.35 | learning rate: 2.624E-05 | global batch size: 256 | lm loss: 3.336073E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 741.875 | TFLOPs: 25.97 | 7: iteration 19020/ 21553 | consumed samples: 4869120 | consumed tokens: 9971957760 | elapsed time per iteration (s): 0.30 | learning rate: 2.619E-05 | global batch size: 256 | lm loss: 3.336785E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.075 | TFLOPs: 30.32 | 7: iteration 19030/ 21553 | consumed samples: 4871680 | consumed tokens: 9977200640 | elapsed time per iteration (s): 0.30 | learning rate: 2.614E-05 | global batch size: 256 | lm loss: 3.334449E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.982 | TFLOPs: 30.32 | 7: iteration 19040/ 21553 | consumed samples: 4874240 | consumed tokens: 9982443520 | elapsed time per iteration (s): 0.30 | learning rate: 2.609E-05 | global batch size: 256 | lm loss: 3.335006E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.750 | TFLOPs: 30.31 | 7: iteration 19050/ 21553 | consumed samples: 4876800 | consumed tokens: 9987686400 | elapsed time per iteration (s): 0.30 | learning rate: 2.604E-05 | global batch size: 256 | lm loss: 3.336412E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.906 | TFLOPs: 30.31 | 7: iteration 19060/ 21553 | consumed samples: 4879360 | consumed tokens: 9992929280 | elapsed time per iteration (s): 0.30 | learning rate: 2.600E-05 | global batch size: 256 | lm loss: 3.338448E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.486 | TFLOPs: 30.33 | 7: iteration 19070/ 21553 | consumed samples: 4881920 | consumed tokens: 9998172160 | elapsed time per iteration (s): 0.30 | learning rate: 2.595E-05 | global batch size: 256 | lm loss: 3.337462E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.621 | TFLOPs: 30.30 | 7: iteration 19080/ 21553 | consumed samples: 4884480 | consumed tokens: 10003415040 | elapsed time per iteration (s): 0.30 | learning rate: 2.590E-05 | global batch size: 256 | lm loss: 3.336013E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.636 | TFLOPs: 30.34 | 7: iteration 19090/ 21553 | consumed samples: 4887040 | consumed tokens: 10008657920 | elapsed time per iteration (s): 0.30 | learning rate: 2.585E-05 | global batch size: 256 | lm loss: 3.337267E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.937 | TFLOPs: 30.31 | 7: iteration 19100/ 21553 | consumed samples: 4889600 | consumed tokens: 10013900800 | elapsed time per iteration (s): 0.30 | learning rate: 2.581E-05 | global batch size: 256 | lm loss: 3.333873E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.978 | TFLOPs: 30.32 | 7: iteration 19110/ 21553 | consumed samples: 4892160 | consumed tokens: 10019143680 | elapsed time per iteration (s): 0.30 | learning rate: 2.576E-05 | global batch size: 256 | lm loss: 3.338376E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.110 | TFLOPs: 30.32 | 7: iteration 19120/ 21553 | consumed samples: 4894720 | consumed tokens: 10024386560 | elapsed time per iteration (s): 0.30 | learning rate: 2.571E-05 | global batch size: 256 | lm loss: 3.331796E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.244 | TFLOPs: 30.32 | 7: iteration 19130/ 21553 | consumed samples: 4897280 | consumed tokens: 10029629440 | elapsed time per iteration (s): 0.30 | learning rate: 2.567E-05 | global batch size: 256 | lm loss: 3.334641E+00 | grad norm: 0.438 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.908 | TFLOPs: 30.31 | 7: iteration 19140/ 21553 | consumed samples: 4899840 | consumed tokens: 10034872320 | elapsed time per iteration (s): 0.30 | learning rate: 2.562E-05 | global batch size: 256 | lm loss: 3.333454E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.186 | TFLOPs: 30.32 | 7: iteration 19150/ 21553 | consumed samples: 4902400 | consumed tokens: 10040115200 | elapsed time per iteration (s): 0.30 | learning rate: 2.557E-05 | global batch size: 256 | lm loss: 3.329887E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.675 | TFLOPs: 30.34 | 7: iteration 19160/ 21553 | consumed samples: 4904960 | consumed tokens: 10045358080 | elapsed time per iteration (s): 0.30 | learning rate: 2.553E-05 | global batch size: 256 | lm loss: 3.340154E+00 | grad norm: 0.402 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.556 | TFLOPs: 30.34 | 7: iteration 19170/ 21553 | consumed samples: 4907520 | consumed tokens: 10050600960 | elapsed time per iteration (s): 0.30 | learning rate: 2.548E-05 | global batch size: 256 | lm loss: 3.337719E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.806 | TFLOPs: 30.34 | 7: iteration 19180/ 21553 | consumed samples: 4910080 | consumed tokens: 10055843840 | elapsed time per iteration (s): 0.30 | learning rate: 2.544E-05 | global batch size: 256 | lm loss: 3.334684E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.609 | TFLOPs: 30.34 | 7: iteration 19190/ 21553 | consumed samples: 4912640 | consumed tokens: 10061086720 | elapsed time per iteration (s): 0.30 | learning rate: 2.539E-05 | global batch size: 256 | lm loss: 3.334183E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.041 | TFLOPs: 30.35 | 7: iteration 19200/ 21553 | consumed samples: 4915200 | consumed tokens: 10066329600 | elapsed time per iteration (s): 0.30 | learning rate: 2.535E-05 | global batch size: 256 | lm loss: 3.335356E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.791 | TFLOPs: 30.34 | 7: iteration 19210/ 21553 | consumed samples: 4917760 | consumed tokens: 10071572480 | elapsed time per iteration (s): 0.30 | learning rate: 2.530E-05 | global batch size: 256 | lm loss: 3.330478E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.492 | TFLOPs: 30.33 | 7: iteration 19220/ 21553 | consumed samples: 4920320 | consumed tokens: 10076815360 | elapsed time per iteration (s): 0.30 | learning rate: 2.526E-05 | global batch size: 256 | lm loss: 3.336878E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.398 | TFLOPs: 30.33 | 7: iteration 19230/ 21553 | consumed samples: 4922880 | consumed tokens: 10082058240 | elapsed time per iteration (s): 0.30 | learning rate: 2.521E-05 | global batch size: 256 | lm loss: 3.339955E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.944 | TFLOPs: 30.31 | 7: iteration 19240/ 21553 | consumed samples: 4925440 | consumed tokens: 10087301120 | elapsed time per iteration (s): 0.30 | learning rate: 2.517E-05 | global batch size: 256 | lm loss: 3.336666E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.921 | TFLOPs: 30.24 | 7: iteration 19250/ 21553 | consumed samples: 4928000 | consumed tokens: 10092544000 | elapsed time per iteration (s): 0.30 | learning rate: 2.512E-05 | global batch size: 256 | lm loss: 3.335624E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.487 | TFLOPs: 30.30 | 7: iteration 19260/ 21553 | consumed samples: 4930560 | consumed tokens: 10097786880 | elapsed time per iteration (s): 0.30 | learning rate: 2.508E-05 | global batch size: 256 | lm loss: 3.338527E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.769 | TFLOPs: 30.31 | 7: iteration 19270/ 21553 | consumed samples: 4933120 | consumed tokens: 10103029760 | elapsed time per iteration (s): 0.30 | learning rate: 2.504E-05 | global batch size: 256 | lm loss: 3.333470E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.608 | TFLOPs: 30.30 | 7: iteration 19280/ 21553 | consumed samples: 4935680 | consumed tokens: 10108272640 | elapsed time per iteration (s): 0.30 | learning rate: 2.499E-05 | global batch size: 256 | lm loss: 3.329466E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.773 | TFLOPs: 30.31 | 7: iteration 19290/ 21553 | consumed samples: 4938240 | consumed tokens: 10113515520 | elapsed time per iteration (s): 0.30 | learning rate: 2.495E-05 | global batch size: 256 | lm loss: 3.333592E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.647 | TFLOPs: 30.30 | 7: iteration 19300/ 21553 | consumed samples: 4940800 | consumed tokens: 10118758400 | elapsed time per iteration (s): 0.30 | learning rate: 2.491E-05 | global batch size: 256 | lm loss: 3.333727E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.073 | TFLOPs: 30.32 | 7: iteration 19310/ 21553 | consumed samples: 4943360 | consumed tokens: 10124001280 | elapsed time per iteration (s): 0.30 | learning rate: 2.486E-05 | global batch size: 256 | lm loss: 3.333619E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.560 | TFLOPs: 30.30 | 7: iteration 19320/ 21553 | consumed samples: 4945920 | consumed tokens: 10129244160 | elapsed time per iteration (s): 0.30 | learning rate: 2.482E-05 | global batch size: 256 | lm loss: 3.332321E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.049 | TFLOPs: 30.32 | 7: iteration 19330/ 21553 | consumed samples: 4948480 | consumed tokens: 10134487040 | elapsed time per iteration (s): 0.30 | learning rate: 2.478E-05 | global batch size: 256 | lm loss: 3.328606E+00 | grad norm: 0.403 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.997 | TFLOPs: 30.32 | 7: iteration 19340/ 21553 | consumed samples: 4951040 | consumed tokens: 10139729920 | elapsed time per iteration (s): 0.30 | learning rate: 2.474E-05 | global batch size: 256 | lm loss: 3.337630E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.036 | TFLOPs: 30.32 | 7: iteration 19350/ 21553 | consumed samples: 4953600 | consumed tokens: 10144972800 | elapsed time per iteration (s): 0.30 | learning rate: 2.469E-05 | global batch size: 256 | lm loss: 3.336621E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.102 | TFLOPs: 30.32 | 7: iteration 19360/ 21553 | consumed samples: 4956160 | consumed tokens: 10150215680 | elapsed time per iteration (s): 0.30 | learning rate: 2.465E-05 | global batch size: 256 | lm loss: 3.333398E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.208 | TFLOPs: 30.32 | 7: iteration 19370/ 21553 | consumed samples: 4958720 | consumed tokens: 10155458560 | elapsed time per iteration (s): 0.30 | learning rate: 2.461E-05 | global batch size: 256 | lm loss: 3.335437E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.393 | TFLOPs: 30.33 | 7: iteration 19380/ 21553 | consumed samples: 4961280 | consumed tokens: 10160701440 | elapsed time per iteration (s): 0.30 | learning rate: 2.457E-05 | global batch size: 256 | lm loss: 3.336742E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.255 | TFLOPs: 30.33 | 7: iteration 19390/ 21553 | consumed samples: 4963840 | consumed tokens: 10165944320 | elapsed time per iteration (s): 0.30 | learning rate: 2.453E-05 | global batch size: 256 | lm loss: 3.333525E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.328 | TFLOPs: 30.33 | 7: iteration 19400/ 21553 | consumed samples: 4966400 | consumed tokens: 10171187200 | elapsed time per iteration (s): 0.30 | learning rate: 2.448E-05 | global batch size: 256 | lm loss: 3.328373E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.882 | TFLOPs: 30.31 | 7: iteration 19410/ 21553 | consumed samples: 4968960 | consumed tokens: 10176430080 | elapsed time per iteration (s): 0.30 | learning rate: 2.444E-05 | global batch size: 256 | lm loss: 3.332310E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.219 | TFLOPs: 30.32 | 7: iteration 19420/ 21553 | consumed samples: 4971520 | consumed tokens: 10181672960 | elapsed time per iteration (s): 0.30 | learning rate: 2.440E-05 | global batch size: 256 | lm loss: 3.336448E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.720 | TFLOPs: 30.34 | 7: iteration 19430/ 21553 | consumed samples: 4974080 | consumed tokens: 10186915840 | elapsed time per iteration (s): 0.30 | learning rate: 2.436E-05 | global batch size: 256 | lm loss: 3.333741E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.638 | TFLOPs: 30.34 | 7: iteration 19440/ 21553 | consumed samples: 4976640 | consumed tokens: 10192158720 | elapsed time per iteration (s): 0.30 | learning rate: 2.432E-05 | global batch size: 256 | lm loss: 3.335174E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.061 | TFLOPs: 30.32 | 7: iteration 19450/ 21553 | consumed samples: 4979200 | consumed tokens: 10197401600 | elapsed time per iteration (s): 0.30 | learning rate: 2.428E-05 | global batch size: 256 | lm loss: 3.335115E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.155 | TFLOPs: 30.32 | 7: iteration 19460/ 21553 | consumed samples: 4981760 | consumed tokens: 10202644480 | elapsed time per iteration (s): 0.30 | learning rate: 2.424E-05 | global batch size: 256 | lm loss: 3.334041E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.032 | TFLOPs: 30.32 | 7: iteration 19470/ 21553 | consumed samples: 4984320 | consumed tokens: 10207887360 | elapsed time per iteration (s): 0.30 | learning rate: 2.420E-05 | global batch size: 256 | lm loss: 3.330392E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.315 | TFLOPs: 30.33 | 7: iteration 19480/ 21553 | consumed samples: 4986880 | consumed tokens: 10213130240 | elapsed time per iteration (s): 0.30 | learning rate: 2.416E-05 | global batch size: 256 | lm loss: 3.327092E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.957 | TFLOPs: 30.31 | 7: iteration 19490/ 21553 | consumed samples: 4989440 | consumed tokens: 10218373120 | elapsed time per iteration (s): 0.30 | learning rate: 2.412E-05 | global batch size: 256 | lm loss: 3.334634E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.150 | TFLOPs: 30.32 | 7: iteration 19500/ 21553 | consumed samples: 4992000 | consumed tokens: 10223616000 | elapsed time per iteration (s): 0.30 | learning rate: 2.408E-05 | global batch size: 256 | lm loss: 3.327968E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.036 | TFLOPs: 30.32 | 7: iteration 19510/ 21553 | consumed samples: 4994560 | consumed tokens: 10228858880 | elapsed time per iteration (s): 0.30 | learning rate: 2.404E-05 | global batch size: 256 | lm loss: 3.335374E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.907 | TFLOPs: 30.31 | 7: iteration 19520/ 21553 | consumed samples: 4997120 | consumed tokens: 10234101760 | elapsed time per iteration (s): 0.30 | learning rate: 2.400E-05 | global batch size: 256 | lm loss: 3.337419E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 856.826 | TFLOPs: 30.00 | 7: iteration 19530/ 21553 | consumed samples: 4999680 | consumed tokens: 10239344640 | elapsed time per iteration (s): 0.30 | learning rate: 2.396E-05 | global batch size: 256 | lm loss: 3.332181E+00 | grad norm: 0.452 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.294 | TFLOPs: 30.33 | 7: iteration 19540/ 21553 | consumed samples: 5002240 | consumed tokens: 10244587520 | elapsed time per iteration (s): 0.30 | learning rate: 2.392E-05 | global batch size: 256 | lm loss: 3.334653E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.194 | TFLOPs: 30.32 | 7: iteration 19550/ 21553 | consumed samples: 5004800 | consumed tokens: 10249830400 | elapsed time per iteration (s): 0.30 | learning rate: 2.389E-05 | global batch size: 256 | lm loss: 3.330283E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.522 | TFLOPs: 30.33 | 7: iteration 19560/ 21553 | consumed samples: 5007360 | consumed tokens: 10255073280 | elapsed time per iteration (s): 0.30 | learning rate: 2.385E-05 | global batch size: 256 | lm loss: 3.332155E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.835 | TFLOPs: 30.35 | 7: iteration 19570/ 21553 | consumed samples: 5009920 | consumed tokens: 10260316160 | elapsed time per iteration (s): 0.30 | learning rate: 2.381E-05 | global batch size: 256 | lm loss: 3.329642E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.018 | TFLOPs: 30.32 | 7: iteration 19580/ 21553 | consumed samples: 5012480 | consumed tokens: 10265559040 | elapsed time per iteration (s): 0.30 | learning rate: 2.377E-05 | global batch size: 256 | lm loss: 3.334989E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.899 | TFLOPs: 30.31 | 7: iteration 19590/ 21553 | consumed samples: 5015040 | consumed tokens: 10270801920 | elapsed time per iteration (s): 0.30 | learning rate: 2.373E-05 | global batch size: 256 | lm loss: 3.330766E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.065 | TFLOPs: 30.32 | 7: iteration 19600/ 21553 | consumed samples: 5017600 | consumed tokens: 10276044800 | elapsed time per iteration (s): 0.30 | learning rate: 2.370E-05 | global batch size: 256 | lm loss: 3.325933E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.332 | TFLOPs: 30.33 | 7: iteration 19610/ 21553 | consumed samples: 5020160 | consumed tokens: 10281287680 | elapsed time per iteration (s): 0.30 | learning rate: 2.366E-05 | global batch size: 256 | lm loss: 3.331924E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.831 | TFLOPs: 30.31 | 7: iteration 19620/ 21553 | consumed samples: 5022720 | consumed tokens: 10286530560 | elapsed time per iteration (s): 0.30 | learning rate: 2.362E-05 | global batch size: 256 | lm loss: 3.339437E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.889 | TFLOPs: 30.31 | 7: iteration 19630/ 21553 | consumed samples: 5025280 | consumed tokens: 10291773440 | elapsed time per iteration (s): 0.30 | learning rate: 2.358E-05 | global batch size: 256 | lm loss: 3.337600E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.924 | TFLOPs: 30.31 | 7: iteration 19640/ 21553 | consumed samples: 5027840 | consumed tokens: 10297016320 | elapsed time per iteration (s): 0.30 | learning rate: 2.355E-05 | global batch size: 256 | lm loss: 3.334957E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.079 | TFLOPs: 30.32 | 7: iteration 19650/ 21553 | consumed samples: 5030400 | consumed tokens: 10302259200 | elapsed time per iteration (s): 0.30 | learning rate: 2.351E-05 | global batch size: 256 | lm loss: 3.327778E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.053 | TFLOPs: 30.32 | 7: iteration 19660/ 21553 | consumed samples: 5032960 | consumed tokens: 10307502080 | elapsed time per iteration (s): 0.30 | learning rate: 2.347E-05 | global batch size: 256 | lm loss: 3.333777E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.275 | TFLOPs: 30.33 | 7: iteration 19670/ 21553 | consumed samples: 5035520 | consumed tokens: 10312744960 | elapsed time per iteration (s): 0.30 | learning rate: 2.344E-05 | global batch size: 256 | lm loss: 3.328983E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.950 | TFLOPs: 30.31 | 7: iteration 19680/ 21553 | consumed samples: 5038080 | consumed tokens: 10317987840 | elapsed time per iteration (s): 0.30 | learning rate: 2.340E-05 | global batch size: 256 | lm loss: 3.329486E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.429 | TFLOPs: 30.30 | 7: iteration 19690/ 21553 | consumed samples: 5040640 | consumed tokens: 10323230720 | elapsed time per iteration (s): 0.30 | learning rate: 2.336E-05 | global batch size: 256 | lm loss: 3.330618E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.060 | TFLOPs: 30.32 | 7: iteration 19700/ 21553 | consumed samples: 5043200 | consumed tokens: 10328473600 | elapsed time per iteration (s): 0.30 | learning rate: 2.333E-05 | global batch size: 256 | lm loss: 3.328192E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.705 | TFLOPs: 30.34 | 7: iteration 19710/ 21553 | consumed samples: 5045760 | consumed tokens: 10333716480 | elapsed time per iteration (s): 0.30 | learning rate: 2.329E-05 | global batch size: 256 | lm loss: 3.329712E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.517 | TFLOPs: 30.33 | 7: iteration 19720/ 21553 | consumed samples: 5048320 | consumed tokens: 10338959360 | elapsed time per iteration (s): 0.30 | learning rate: 2.326E-05 | global batch size: 256 | lm loss: 3.330059E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.721 | TFLOPs: 30.34 | 7: iteration 19730/ 21553 | consumed samples: 5050880 | consumed tokens: 10344202240 | elapsed time per iteration (s): 0.30 | learning rate: 2.322E-05 | global batch size: 256 | lm loss: 3.330598E+00 | grad norm: 0.408 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.414 | TFLOPs: 30.33 | 7: iteration 19740/ 21553 | consumed samples: 5053440 | consumed tokens: 10349445120 | elapsed time per iteration (s): 0.30 | learning rate: 2.319E-05 | global batch size: 256 | lm loss: 3.338974E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.645 | TFLOPs: 30.34 | 7: iteration 19750/ 21553 | consumed samples: 5056000 | consumed tokens: 10354688000 | elapsed time per iteration (s): 0.30 | learning rate: 2.315E-05 | global batch size: 256 | lm loss: 3.326468E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.001 | TFLOPs: 30.32 | 7: iteration 19760/ 21553 | consumed samples: 5058560 | consumed tokens: 10359930880 | elapsed time per iteration (s): 0.30 | learning rate: 2.312E-05 | global batch size: 256 | lm loss: 3.326113E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.418 | TFLOPs: 30.33 | 7: iteration 19770/ 21553 | consumed samples: 5061120 | consumed tokens: 10365173760 | elapsed time per iteration (s): 0.30 | learning rate: 2.308E-05 | global batch size: 256 | lm loss: 3.336216E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.403 | TFLOPs: 30.33 | 7: iteration 19780/ 21553 | consumed samples: 5063680 | consumed tokens: 10370416640 | elapsed time per iteration (s): 0.30 | learning rate: 2.305E-05 | global batch size: 256 | lm loss: 3.335999E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.596 | TFLOPs: 30.34 | 7: iteration 19790/ 21553 | consumed samples: 5066240 | consumed tokens: 10375659520 | elapsed time per iteration (s): 0.30 | learning rate: 2.302E-05 | global batch size: 256 | lm loss: 3.328179E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.254 | TFLOPs: 30.33 | 7: iteration 19800/ 21553 | consumed samples: 5068800 | consumed tokens: 10380902400 | elapsed time per iteration (s): 0.30 | learning rate: 2.298E-05 | global batch size: 256 | lm loss: 3.333288E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.455 | TFLOPs: 30.33 | 7: iteration 19810/ 21553 | consumed samples: 5071360 | consumed tokens: 10386145280 | elapsed time per iteration (s): 0.30 | learning rate: 2.295E-05 | global batch size: 256 | lm loss: 3.331535E+00 | grad norm: 0.451 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.800 | TFLOPs: 30.31 | 7: iteration 19820/ 21553 | consumed samples: 5073920 | consumed tokens: 10391388160 | elapsed time per iteration (s): 0.30 | learning rate: 2.291E-05 | global batch size: 256 | lm loss: 3.327603E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.417 | TFLOPs: 30.33 | 7: iteration 19830/ 21553 | consumed samples: 5076480 | consumed tokens: 10396631040 | elapsed time per iteration (s): 0.30 | learning rate: 2.288E-05 | global batch size: 256 | lm loss: 3.330647E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.624 | TFLOPs: 30.34 | 7: iteration 19840/ 21553 | consumed samples: 5079040 | consumed tokens: 10401873920 | elapsed time per iteration (s): 0.30 | learning rate: 2.285E-05 | global batch size: 256 | lm loss: 3.332187E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.326 | TFLOPs: 30.33 | 7: iteration 19850/ 21553 | consumed samples: 5081600 | consumed tokens: 10407116800 | elapsed time per iteration (s): 0.30 | learning rate: 2.281E-05 | global batch size: 256 | lm loss: 3.334453E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.229 | TFLOPs: 30.32 | 7: iteration 19860/ 21553 | consumed samples: 5084160 | consumed tokens: 10412359680 | elapsed time per iteration (s): 0.30 | learning rate: 2.278E-05 | global batch size: 256 | lm loss: 3.327924E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.211 | TFLOPs: 30.32 | 7: iteration 19870/ 21553 | consumed samples: 5086720 | consumed tokens: 10417602560 | elapsed time per iteration (s): 0.30 | learning rate: 2.275E-05 | global batch size: 256 | lm loss: 3.332340E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.569 | TFLOPs: 30.34 | 7: iteration 19880/ 21553 | consumed samples: 5089280 | consumed tokens: 10422845440 | elapsed time per iteration (s): 0.30 | learning rate: 2.272E-05 | global batch size: 256 | lm loss: 3.340667E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.646 | TFLOPs: 30.34 | 7: iteration 19890/ 21553 | consumed samples: 5091840 | consumed tokens: 10428088320 | elapsed time per iteration (s): 0.30 | learning rate: 2.268E-05 | global batch size: 256 | lm loss: 3.333088E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.472 | TFLOPs: 30.33 | 7: iteration 19900/ 21553 | consumed samples: 5094400 | consumed tokens: 10433331200 | elapsed time per iteration (s): 0.30 | learning rate: 2.265E-05 | global batch size: 256 | lm loss: 3.331309E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.566 | TFLOPs: 30.34 | 7: iteration 19910/ 21553 | consumed samples: 5096960 | consumed tokens: 10438574080 | elapsed time per iteration (s): 0.30 | learning rate: 2.262E-05 | global batch size: 256 | lm loss: 3.325328E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.102 | TFLOPs: 30.32 | 7: iteration 19920/ 21553 | consumed samples: 5099520 | consumed tokens: 10443816960 | elapsed time per iteration (s): 0.30 | learning rate: 2.259E-05 | global batch size: 256 | lm loss: 3.331678E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.057 | TFLOPs: 30.32 | 7: iteration 19930/ 21553 | consumed samples: 5102080 | consumed tokens: 10449059840 | elapsed time per iteration (s): 0.30 | learning rate: 2.256E-05 | global batch size: 256 | lm loss: 3.330783E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.053 | TFLOPs: 30.32 | 7: iteration 19940/ 21553 | consumed samples: 5104640 | consumed tokens: 10454302720 | elapsed time per iteration (s): 0.30 | learning rate: 2.253E-05 | global batch size: 256 | lm loss: 3.327299E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.001 | TFLOPs: 30.35 | 7: iteration 19950/ 21553 | consumed samples: 5107200 | consumed tokens: 10459545600 | elapsed time per iteration (s): 0.30 | learning rate: 2.250E-05 | global batch size: 256 | lm loss: 3.330173E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.849 | TFLOPs: 30.35 | 7: iteration 19960/ 21553 | consumed samples: 5109760 | consumed tokens: 10464788480 | elapsed time per iteration (s): 0.30 | learning rate: 2.246E-05 | global batch size: 256 | lm loss: 3.339856E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.288 | TFLOPs: 30.33 | 7: iteration 19970/ 21553 | consumed samples: 5112320 | consumed tokens: 10470031360 | elapsed time per iteration (s): 0.30 | learning rate: 2.243E-05 | global batch size: 256 | lm loss: 3.328653E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.854 | TFLOPs: 30.35 | 7: iteration 19980/ 21553 | consumed samples: 5114880 | consumed tokens: 10475274240 | elapsed time per iteration (s): 0.30 | learning rate: 2.240E-05 | global batch size: 256 | lm loss: 3.334521E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.155 | TFLOPs: 30.36 | 7: iteration 19990/ 21553 | consumed samples: 5117440 | consumed tokens: 10480517120 | elapsed time per iteration (s): 0.30 | learning rate: 2.237E-05 | global batch size: 256 | lm loss: 3.332777E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 867.040 | TFLOPs: 30.35 | 0: [2023-03-17 01:35:02,069] [INFO] [logging.py:68:log_dist] [Rank 0] step=20000, skipped=0, lr=[2.234259142486404e-05, 2.234259142486404e-05, 2.234259142486404e-05], mom=[(0.9, 0.999), (0.9, 0.999), (0.9, 0.999)] 7: iteration 20000/ 21553 | consumed samples: 5120000 | consumed tokens: 10485760000 | elapsed time per iteration (s): 0.30 | learning rate: 2.234E-05 | global batch size: 256 | lm loss: 3.331540E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.950 | TFLOPs: 30.35 | 0: steps: 20000 loss: 3.3448 iter time (s): 0.294 samples/sec: 870.213 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 20000 | lm loss value: 3.965218E+00 | lm loss PPL: 5.273177E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 20000 to checkpoints_146m14b100mdedup 0: [2023-03-17 01:35:02,190] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step20000 is begin to save! 0: [2023-03-17 01:35:02,193] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:35:02,280] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:35:02,281] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:35:02,297] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:35:02,297] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:35:02,312] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:35:02,312] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:35:02,327] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:35:02,327] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:35:02,342] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:35:02,342] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:35:02,357] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:35:02,357] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:35:02,372] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:35:02,372] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:35:02,387] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:35:02,387] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:35:02,401] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:35:02,402] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:35:02,416] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:35:02,416] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:35:02,431] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:35:02,431] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:35:02,446] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:35:02,446] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:35:02,461] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:35:02,461] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:35:02,476] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:35:02,476] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:35:02,490] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:35:02,491] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:35:02,505] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:35:02,505] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:35:02,507] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:35:02,507] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step20000/mp_rank_00_model_states.pt 0: [2023-03-17 01:35:02,507] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/mp_rank_00_model_states.pt... 0: [2023-03-17 01:35:02,509] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/mp_rank_00_model_states.pt. 0: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:35:02,526] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:35:02,558] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:35:02,558] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 01:35:02,558] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,560] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,560] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,560] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 01:35:02,560] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 0: [2023-03-17 01:35:02,561] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,561] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:35:02,561] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 0: [2023-03-17 01:35:02,561] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,562] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,562] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 01:35:02,562] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 3: [2023-03-17 01:35:02,562] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,562] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:35:02,562] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,560] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 01:35:02,560] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 4: [2023-03-17 01:35:02,563] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:35:02,563] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 01:35:02,563] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 7: [2023-03-17 01:35:02,562] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:35:02,563] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:35:02,563] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:35:02,562] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 01:35:02,562] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 7: [2023-03-17 01:35:02,562] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:35:02,563] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 01:35:02,563] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 7: [2023-03-17 01:35:02,562] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 01:35:02,562] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 4: [2023-03-17 01:35:02,563] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 4: [2023-03-17 01:35:02,563] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,563] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,564] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 3: [2023-03-17 01:35:02,564] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 0: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 5: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 4: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:35:02,564] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 4: [2023-03-17 01:35:02,564] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 5: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 4: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 3: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,564] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 5: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:35:02,565] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 5: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 6: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,565] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 1: [2023-03-17 01:35:02,566] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:35:02,566] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:35:02,566] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:35:02,566] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 01:35:02,566] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 1: [2023-03-17 01:35:02,566] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 6: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:35:02,565] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 01:35:02,565] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 6: [2023-03-17 01:35:02,565] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 6: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 3: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:35:02,565] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:35:02,565] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 01:35:02,566] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 3: [2023-03-17 01:35:02,567] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 6: [2023-03-17 01:35:02,566] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 1: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 1: [2023-03-17 01:35:02,567] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 5: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:35:02,567] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 1: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 5: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 5: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:35:02,567] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 3: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,567] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 01:35:02,568] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 5: [2023-03-17 01:35:02,568] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:35:02,568] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 5: [2023-03-17 01:35:02,568] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 3: [2023-03-17 01:35:02,568] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,568] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 01:35:02,568] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,563] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:35:02,563] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:35:02,564] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,566] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:35:02,566] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 01:35:02,566] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:35:02,567] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:35:02,567] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:35:02,567] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 4: [2023-03-17 01:35:02,569] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:35:02,569] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:35:02,569] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 01:35:02,569] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:35:02,569] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 4: [2023-03-17 01:35:02,569] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 4: [2023-03-17 01:35:02,569] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:35:02,569] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 01:35:02,569] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 6: [2023-03-17 01:35:02,569] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:35:02,569] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 01:35:02,569] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 6: [2023-03-17 01:35:02,569] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:35:02,569] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 3: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,570] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 6: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 5: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:35:02,570] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 1: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 1: [2023-03-17 01:35:02,570] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 0: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:35:02,570] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 01:35:02,570] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 0: [2023-03-17 01:35:02,570] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 0: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 0: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 6: [2023-03-17 01:35:02,570] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 2: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,570] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:35:02,570] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 2: [2023-03-17 01:35:02,571] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:35:02,571] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 01:35:02,571] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 5: [2023-03-17 01:35:02,571] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:35:02,571] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:35:02,571] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 01:35:02,571] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 01:35:02,571] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 5: [2023-03-17 01:35:02,571] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 7: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:35:02,564] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 7: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:35:02,564] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 01:35:02,564] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 7: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:35:02,565] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 01:35:02,565] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 7: [2023-03-17 01:35:02,573] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:35:02,573] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:35:02,573] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:35:02,573] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:35:02,573] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 01:35:02,573] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 7: [2023-03-17 01:35:02,573] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 7: [2023-03-17 01:35:02,573] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:35:02,573] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 1: [2023-03-17 01:35:02,576] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:35:02,576] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:35:02,576] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 01:35:02,576] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:35:02,576] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:35:02,576] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:35:02,576] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 1: [2023-03-17 01:35:02,577] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 1: [2023-03-17 01:35:02,576] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 01:35:02,576] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 01:35:02,577] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 1: [2023-03-17 01:35:02,577] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 0: [2023-03-17 01:35:02,586] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step20000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:35:02,586] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step20000 is ready now! 0: successfully saved checkpoint at iteration 20000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 401.74 7: iteration 20010/ 21553 | consumed samples: 5122560 | consumed tokens: 10491002880 | elapsed time per iteration (s): 0.35 | learning rate: 2.231E-05 | global batch size: 256 | lm loss: 3.328335E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 738.150 | TFLOPs: 25.84 | 7: iteration 20020/ 21553 | consumed samples: 5125120 | consumed tokens: 10496245760 | elapsed time per iteration (s): 0.30 | learning rate: 2.228E-05 | global batch size: 256 | lm loss: 3.326512E+00 | grad norm: 0.409 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.193 | TFLOPs: 30.32 | 7: iteration 20030/ 21553 | consumed samples: 5127680 | consumed tokens: 10501488640 | elapsed time per iteration (s): 0.30 | learning rate: 2.225E-05 | global batch size: 256 | lm loss: 3.328935E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.428 | TFLOPs: 30.30 | 7: iteration 20040/ 21553 | consumed samples: 5130240 | consumed tokens: 10506731520 | elapsed time per iteration (s): 0.30 | learning rate: 2.222E-05 | global batch size: 256 | lm loss: 3.325552E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.120 | TFLOPs: 30.32 | 7: iteration 20050/ 21553 | consumed samples: 5132800 | consumed tokens: 10511974400 | elapsed time per iteration (s): 0.30 | learning rate: 2.219E-05 | global batch size: 256 | lm loss: 3.336579E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.597 | TFLOPs: 30.30 | 7: iteration 20060/ 21553 | consumed samples: 5135360 | consumed tokens: 10517217280 | elapsed time per iteration (s): 0.30 | learning rate: 2.217E-05 | global batch size: 256 | lm loss: 3.329180E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.826 | TFLOPs: 30.31 | 7: iteration 20070/ 21553 | consumed samples: 5137920 | consumed tokens: 10522460160 | elapsed time per iteration (s): 0.30 | learning rate: 2.214E-05 | global batch size: 256 | lm loss: 3.330967E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.627 | TFLOPs: 30.30 | 7: iteration 20080/ 21553 | consumed samples: 5140480 | consumed tokens: 10527703040 | elapsed time per iteration (s): 0.30 | learning rate: 2.211E-05 | global batch size: 256 | lm loss: 3.329303E+00 | grad norm: 0.444 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.475 | TFLOPs: 30.30 | 7: iteration 20090/ 21553 | consumed samples: 5143040 | consumed tokens: 10532945920 | elapsed time per iteration (s): 0.30 | learning rate: 2.208E-05 | global batch size: 256 | lm loss: 3.327348E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.532 | TFLOPs: 30.30 | 7: iteration 20100/ 21553 | consumed samples: 5145600 | consumed tokens: 10538188800 | elapsed time per iteration (s): 0.30 | learning rate: 2.205E-05 | global batch size: 256 | lm loss: 3.324406E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.409 | TFLOPs: 30.30 | 7: iteration 20110/ 21553 | consumed samples: 5148160 | consumed tokens: 10543431680 | elapsed time per iteration (s): 0.30 | learning rate: 2.202E-05 | global batch size: 256 | lm loss: 3.324575E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.110 | TFLOPs: 30.29 | 7: iteration 20120/ 21553 | consumed samples: 5150720 | consumed tokens: 10548674560 | elapsed time per iteration (s): 0.30 | learning rate: 2.200E-05 | global batch size: 256 | lm loss: 3.325952E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.793 | TFLOPs: 30.31 | 7: iteration 20130/ 21553 | consumed samples: 5153280 | consumed tokens: 10553917440 | elapsed time per iteration (s): 0.30 | learning rate: 2.197E-05 | global batch size: 256 | lm loss: 3.325168E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.100 | TFLOPs: 30.32 | 7: iteration 20140/ 21553 | consumed samples: 5155840 | consumed tokens: 10559160320 | elapsed time per iteration (s): 0.30 | learning rate: 2.194E-05 | global batch size: 256 | lm loss: 3.328849E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.161 | TFLOPs: 30.32 | 7: iteration 20150/ 21553 | consumed samples: 5158400 | consumed tokens: 10564403200 | elapsed time per iteration (s): 0.30 | learning rate: 2.191E-05 | global batch size: 256 | lm loss: 3.333395E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.102 | TFLOPs: 30.32 | 7: iteration 20160/ 21553 | consumed samples: 5160960 | consumed tokens: 10569646080 | elapsed time per iteration (s): 0.30 | learning rate: 2.189E-05 | global batch size: 256 | lm loss: 3.333451E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.015 | TFLOPs: 30.32 | 7: iteration 20170/ 21553 | consumed samples: 5163520 | consumed tokens: 10574888960 | elapsed time per iteration (s): 0.30 | learning rate: 2.186E-05 | global batch size: 256 | lm loss: 3.325758E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.884 | TFLOPs: 30.31 | 7: iteration 20180/ 21553 | consumed samples: 5166080 | consumed tokens: 10580131840 | elapsed time per iteration (s): 0.30 | learning rate: 2.183E-05 | global batch size: 256 | lm loss: 3.324120E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.234 | TFLOPs: 30.32 | 7: iteration 20190/ 21553 | consumed samples: 5168640 | consumed tokens: 10585374720 | elapsed time per iteration (s): 0.30 | learning rate: 2.181E-05 | global batch size: 256 | lm loss: 3.323544E+00 | grad norm: 0.445 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.165 | TFLOPs: 30.32 | 7: iteration 20200/ 21553 | consumed samples: 5171200 | consumed tokens: 10590617600 | elapsed time per iteration (s): 0.30 | learning rate: 2.178E-05 | global batch size: 256 | lm loss: 3.319893E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.973 | TFLOPs: 30.32 | 7: iteration 20210/ 21553 | consumed samples: 5173760 | consumed tokens: 10595860480 | elapsed time per iteration (s): 0.30 | learning rate: 2.175E-05 | global batch size: 256 | lm loss: 3.328486E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.198 | TFLOPs: 30.32 | 7: iteration 20220/ 21553 | consumed samples: 5176320 | consumed tokens: 10601103360 | elapsed time per iteration (s): 0.30 | learning rate: 2.173E-05 | global batch size: 256 | lm loss: 3.328517E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.321 | TFLOPs: 30.33 | 7: iteration 20230/ 21553 | consumed samples: 5178880 | consumed tokens: 10606346240 | elapsed time per iteration (s): 0.30 | learning rate: 2.170E-05 | global batch size: 256 | lm loss: 3.325763E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.612 | TFLOPs: 30.30 | 7: iteration 20240/ 21553 | consumed samples: 5181440 | consumed tokens: 10611589120 | elapsed time per iteration (s): 0.30 | learning rate: 2.168E-05 | global batch size: 256 | lm loss: 3.319904E+00 | grad norm: 0.412 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.394 | TFLOPs: 30.33 | 7: iteration 20250/ 21553 | consumed samples: 5184000 | consumed tokens: 10616832000 | elapsed time per iteration (s): 0.30 | learning rate: 2.165E-05 | global batch size: 256 | lm loss: 3.330433E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.361 | TFLOPs: 30.33 | 7: iteration 20260/ 21553 | consumed samples: 5186560 | consumed tokens: 10622074880 | elapsed time per iteration (s): 0.30 | learning rate: 2.163E-05 | global batch size: 256 | lm loss: 3.334753E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.446 | TFLOPs: 30.30 | 7: iteration 20270/ 21553 | consumed samples: 5189120 | consumed tokens: 10627317760 | elapsed time per iteration (s): 0.30 | learning rate: 2.160E-05 | global batch size: 256 | lm loss: 3.333289E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.907 | TFLOPs: 30.28 | 7: iteration 20280/ 21553 | consumed samples: 5191680 | consumed tokens: 10632560640 | elapsed time per iteration (s): 0.30 | learning rate: 2.158E-05 | global batch size: 256 | lm loss: 3.323162E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.242 | TFLOPs: 30.29 | 7: iteration 20290/ 21553 | consumed samples: 5194240 | consumed tokens: 10637803520 | elapsed time per iteration (s): 0.30 | learning rate: 2.155E-05 | global batch size: 256 | lm loss: 3.330917E+00 | grad norm: 0.411 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.228 | TFLOPs: 30.29 | 7: iteration 20300/ 21553 | consumed samples: 5196800 | consumed tokens: 10643046400 | elapsed time per iteration (s): 0.30 | learning rate: 2.153E-05 | global batch size: 256 | lm loss: 3.320353E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.759 | TFLOPs: 30.27 | 7: iteration 20310/ 21553 | consumed samples: 5199360 | consumed tokens: 10648289280 | elapsed time per iteration (s): 0.30 | learning rate: 2.150E-05 | global batch size: 256 | lm loss: 3.326226E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.430 | TFLOPs: 30.30 | 7: iteration 20320/ 21553 | consumed samples: 5201920 | consumed tokens: 10653532160 | elapsed time per iteration (s): 0.30 | learning rate: 2.148E-05 | global batch size: 256 | lm loss: 3.331794E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.756 | TFLOPs: 30.31 | 7: iteration 20330/ 21553 | consumed samples: 5204480 | consumed tokens: 10658775040 | elapsed time per iteration (s): 0.30 | learning rate: 2.146E-05 | global batch size: 256 | lm loss: 3.327663E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.347 | TFLOPs: 30.29 | 7: iteration 20340/ 21553 | consumed samples: 5207040 | consumed tokens: 10664017920 | elapsed time per iteration (s): 0.30 | learning rate: 2.143E-05 | global batch size: 256 | lm loss: 3.322724E+00 | grad norm: 0.444 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.472 | TFLOPs: 30.30 | 7: iteration 20350/ 21553 | consumed samples: 5209600 | consumed tokens: 10669260800 | elapsed time per iteration (s): 0.30 | learning rate: 2.141E-05 | global batch size: 256 | lm loss: 3.327074E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.195 | TFLOPs: 30.29 | 7: iteration 20360/ 21553 | consumed samples: 5212160 | consumed tokens: 10674503680 | elapsed time per iteration (s): 0.30 | learning rate: 2.138E-05 | global batch size: 256 | lm loss: 3.330722E+00 | grad norm: 0.405 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.914 | TFLOPs: 30.31 | 7: iteration 20370/ 21553 | consumed samples: 5214720 | consumed tokens: 10679746560 | elapsed time per iteration (s): 0.30 | learning rate: 2.136E-05 | global batch size: 256 | lm loss: 3.327189E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.450 | TFLOPs: 30.30 | 7: iteration 20380/ 21553 | consumed samples: 5217280 | consumed tokens: 10684989440 | elapsed time per iteration (s): 0.30 | learning rate: 2.134E-05 | global batch size: 256 | lm loss: 3.325824E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.852 | TFLOPs: 30.31 | 7: iteration 20390/ 21553 | consumed samples: 5219840 | consumed tokens: 10690232320 | elapsed time per iteration (s): 0.30 | learning rate: 2.132E-05 | global batch size: 256 | lm loss: 3.324093E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.670 | TFLOPs: 30.30 | 7: iteration 20400/ 21553 | consumed samples: 5222400 | consumed tokens: 10695475200 | elapsed time per iteration (s): 0.30 | learning rate: 2.129E-05 | global batch size: 256 | lm loss: 3.330782E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.676 | TFLOPs: 30.30 | 7: iteration 20410/ 21553 | consumed samples: 5224960 | consumed tokens: 10700718080 | elapsed time per iteration (s): 0.30 | learning rate: 2.127E-05 | global batch size: 256 | lm loss: 3.326620E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.527 | TFLOPs: 30.30 | 7: iteration 20420/ 21553 | consumed samples: 5227520 | consumed tokens: 10705960960 | elapsed time per iteration (s): 0.30 | learning rate: 2.125E-05 | global batch size: 256 | lm loss: 3.329255E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.757 | TFLOPs: 30.31 | 7: iteration 20430/ 21553 | consumed samples: 5230080 | consumed tokens: 10711203840 | elapsed time per iteration (s): 0.30 | learning rate: 2.123E-05 | global batch size: 256 | lm loss: 3.330512E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.916 | TFLOPs: 30.31 | 7: iteration 20440/ 21553 | consumed samples: 5232640 | consumed tokens: 10716446720 | elapsed time per iteration (s): 0.30 | learning rate: 2.121E-05 | global batch size: 256 | lm loss: 3.323898E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.145 | TFLOPs: 30.32 | 7: iteration 20450/ 21553 | consumed samples: 5235200 | consumed tokens: 10721689600 | elapsed time per iteration (s): 0.30 | learning rate: 2.118E-05 | global batch size: 256 | lm loss: 3.326134E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.794 | TFLOPs: 30.31 | 7: iteration 20460/ 21553 | consumed samples: 5237760 | consumed tokens: 10726932480 | elapsed time per iteration (s): 0.30 | learning rate: 2.116E-05 | global batch size: 256 | lm loss: 3.326133E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.550 | TFLOPs: 30.30 | 7: iteration 20470/ 21553 | consumed samples: 5240320 | consumed tokens: 10732175360 | elapsed time per iteration (s): 0.30 | learning rate: 2.114E-05 | global batch size: 256 | lm loss: 3.320511E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.966 | TFLOPs: 30.32 | 7: iteration 20480/ 21553 | consumed samples: 5242880 | consumed tokens: 10737418240 | elapsed time per iteration (s): 0.30 | learning rate: 2.112E-05 | global batch size: 256 | lm loss: 3.329901E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.256 | TFLOPs: 30.33 | 7: iteration 20490/ 21553 | consumed samples: 5245440 | consumed tokens: 10742661120 | elapsed time per iteration (s): 0.30 | learning rate: 2.110E-05 | global batch size: 256 | lm loss: 3.328164E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.460 | TFLOPs: 30.33 | 7: iteration 20500/ 21553 | consumed samples: 5248000 | consumed tokens: 10747904000 | elapsed time per iteration (s): 0.30 | learning rate: 2.108E-05 | global batch size: 256 | lm loss: 3.332498E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.295 | TFLOPs: 30.33 | 7: iteration 20510/ 21553 | consumed samples: 5250560 | consumed tokens: 10753146880 | elapsed time per iteration (s): 0.30 | learning rate: 2.106E-05 | global batch size: 256 | lm loss: 3.321666E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.934 | TFLOPs: 30.31 | 7: iteration 20520/ 21553 | consumed samples: 5253120 | consumed tokens: 10758389760 | elapsed time per iteration (s): 0.30 | learning rate: 2.104E-05 | global batch size: 256 | lm loss: 3.328514E+00 | grad norm: 0.413 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.188 | TFLOPs: 30.32 | 7: iteration 20530/ 21553 | consumed samples: 5255680 | consumed tokens: 10763632640 | elapsed time per iteration (s): 0.30 | learning rate: 2.102E-05 | global batch size: 256 | lm loss: 3.319955E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.424 | TFLOPs: 30.33 | 7: iteration 20540/ 21553 | consumed samples: 5258240 | consumed tokens: 10768875520 | elapsed time per iteration (s): 0.30 | learning rate: 2.100E-05 | global batch size: 256 | lm loss: 3.324337E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.437 | TFLOPs: 30.33 | 7: iteration 20550/ 21553 | consumed samples: 5260800 | consumed tokens: 10774118400 | elapsed time per iteration (s): 0.30 | learning rate: 2.098E-05 | global batch size: 256 | lm loss: 3.326865E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.104 | TFLOPs: 30.32 | 7: iteration 20560/ 21553 | consumed samples: 5263360 | consumed tokens: 10779361280 | elapsed time per iteration (s): 0.30 | learning rate: 2.096E-05 | global batch size: 256 | lm loss: 3.329366E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.485 | TFLOPs: 30.30 | 7: iteration 20570/ 21553 | consumed samples: 5265920 | consumed tokens: 10784604160 | elapsed time per iteration (s): 0.30 | learning rate: 2.094E-05 | global batch size: 256 | lm loss: 3.327290E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.522 | TFLOPs: 30.26 | 7: iteration 20580/ 21553 | consumed samples: 5268480 | consumed tokens: 10789847040 | elapsed time per iteration (s): 0.30 | learning rate: 2.092E-05 | global batch size: 256 | lm loss: 3.325929E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.306 | TFLOPs: 30.26 | 7: iteration 20590/ 21553 | consumed samples: 5271040 | consumed tokens: 10795089920 | elapsed time per iteration (s): 0.30 | learning rate: 2.090E-05 | global batch size: 256 | lm loss: 3.323812E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.415 | TFLOPs: 30.33 | 7: iteration 20600/ 21553 | consumed samples: 5273600 | consumed tokens: 10800332800 | elapsed time per iteration (s): 0.30 | learning rate: 2.088E-05 | global batch size: 256 | lm loss: 3.325087E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.970 | TFLOPs: 30.28 | 7: iteration 20610/ 21553 | consumed samples: 5276160 | consumed tokens: 10805575680 | elapsed time per iteration (s): 0.30 | learning rate: 2.087E-05 | global batch size: 256 | lm loss: 3.323359E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.168 | TFLOPs: 30.29 | 7: iteration 20620/ 21553 | consumed samples: 5278720 | consumed tokens: 10810818560 | elapsed time per iteration (s): 0.30 | learning rate: 2.085E-05 | global batch size: 256 | lm loss: 3.325510E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.078 | TFLOPs: 30.28 | 7: iteration 20630/ 21553 | consumed samples: 5281280 | consumed tokens: 10816061440 | elapsed time per iteration (s): 0.30 | learning rate: 2.083E-05 | global batch size: 256 | lm loss: 3.317339E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.956 | TFLOPs: 30.28 | 7: iteration 20640/ 21553 | consumed samples: 5283840 | consumed tokens: 10821304320 | elapsed time per iteration (s): 0.30 | learning rate: 2.081E-05 | global batch size: 256 | lm loss: 3.324062E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.365 | TFLOPs: 30.29 | 7: iteration 20650/ 21553 | consumed samples: 5286400 | consumed tokens: 10826547200 | elapsed time per iteration (s): 0.30 | learning rate: 2.079E-05 | global batch size: 256 | lm loss: 3.324685E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.903 | TFLOPs: 30.28 | 7: iteration 20660/ 21553 | consumed samples: 5288960 | consumed tokens: 10831790080 | elapsed time per iteration (s): 0.30 | learning rate: 2.078E-05 | global batch size: 256 | lm loss: 3.324252E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.965 | TFLOPs: 30.28 | 7: iteration 20670/ 21553 | consumed samples: 5291520 | consumed tokens: 10837032960 | elapsed time per iteration (s): 0.30 | learning rate: 2.076E-05 | global batch size: 256 | lm loss: 3.323431E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.927 | TFLOPs: 30.28 | 7: iteration 20680/ 21553 | consumed samples: 5294080 | consumed tokens: 10842275840 | elapsed time per iteration (s): 0.30 | learning rate: 2.074E-05 | global batch size: 256 | lm loss: 3.322448E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.258 | TFLOPs: 30.29 | 7: iteration 20690/ 21553 | consumed samples: 5296640 | consumed tokens: 10847518720 | elapsed time per iteration (s): 0.30 | learning rate: 2.073E-05 | global batch size: 256 | lm loss: 3.324155E+00 | grad norm: 0.435 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.615 | TFLOPs: 30.30 | 7: iteration 20700/ 21553 | consumed samples: 5299200 | consumed tokens: 10852761600 | elapsed time per iteration (s): 0.30 | learning rate: 2.071E-05 | global batch size: 256 | lm loss: 3.318810E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.092 | TFLOPs: 30.28 | 7: iteration 20710/ 21553 | consumed samples: 5301760 | consumed tokens: 10858004480 | elapsed time per iteration (s): 0.30 | learning rate: 2.069E-05 | global batch size: 256 | lm loss: 3.323383E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.262 | TFLOPs: 30.26 | 7: iteration 20720/ 21553 | consumed samples: 5304320 | consumed tokens: 10863247360 | elapsed time per iteration (s): 0.30 | learning rate: 2.068E-05 | global batch size: 256 | lm loss: 3.325096E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 859.403 | TFLOPs: 30.09 | 7: iteration 20730/ 21553 | consumed samples: 5306880 | consumed tokens: 10868490240 | elapsed time per iteration (s): 0.30 | learning rate: 2.066E-05 | global batch size: 256 | lm loss: 3.326144E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.392 | TFLOPs: 30.29 | 7: iteration 20740/ 21553 | consumed samples: 5309440 | consumed tokens: 10873733120 | elapsed time per iteration (s): 0.30 | learning rate: 2.064E-05 | global batch size: 256 | lm loss: 3.325404E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.804 | TFLOPs: 30.27 | 7: iteration 20750/ 21553 | consumed samples: 5312000 | consumed tokens: 10878976000 | elapsed time per iteration (s): 0.30 | learning rate: 2.063E-05 | global batch size: 256 | lm loss: 3.322025E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.673 | TFLOPs: 30.27 | 7: iteration 20760/ 21553 | consumed samples: 5314560 | consumed tokens: 10884218880 | elapsed time per iteration (s): 0.30 | learning rate: 2.061E-05 | global batch size: 256 | lm loss: 3.328487E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.951 | TFLOPs: 30.28 | 7: iteration 20770/ 21553 | consumed samples: 5317120 | consumed tokens: 10889461760 | elapsed time per iteration (s): 0.30 | learning rate: 2.060E-05 | global batch size: 256 | lm loss: 3.325425E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.472 | TFLOPs: 30.26 | 7: iteration 20780/ 21553 | consumed samples: 5319680 | consumed tokens: 10894704640 | elapsed time per iteration (s): 0.30 | learning rate: 2.058E-05 | global batch size: 256 | lm loss: 3.321722E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.291 | TFLOPs: 30.26 | 7: iteration 20790/ 21553 | consumed samples: 5322240 | consumed tokens: 10899947520 | elapsed time per iteration (s): 0.30 | learning rate: 2.057E-05 | global batch size: 256 | lm loss: 3.323567E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.422 | TFLOPs: 30.26 | 7: iteration 20800/ 21553 | consumed samples: 5324800 | consumed tokens: 10905190400 | elapsed time per iteration (s): 0.30 | learning rate: 2.055E-05 | global batch size: 256 | lm loss: 3.331563E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.692 | TFLOPs: 30.27 | 7: iteration 20810/ 21553 | consumed samples: 5327360 | consumed tokens: 10910433280 | elapsed time per iteration (s): 0.30 | learning rate: 2.054E-05 | global batch size: 256 | lm loss: 3.324304E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.630 | TFLOPs: 30.23 | 7: iteration 20820/ 21553 | consumed samples: 5329920 | consumed tokens: 10915676160 | elapsed time per iteration (s): 0.30 | learning rate: 2.052E-05 | global batch size: 256 | lm loss: 3.316452E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.274 | TFLOPs: 30.26 | 7: iteration 20830/ 21553 | consumed samples: 5332480 | consumed tokens: 10920919040 | elapsed time per iteration (s): 0.30 | learning rate: 2.051E-05 | global batch size: 256 | lm loss: 3.320464E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.229 | TFLOPs: 30.25 | 7: iteration 20840/ 21553 | consumed samples: 5335040 | consumed tokens: 10926161920 | elapsed time per iteration (s): 0.30 | learning rate: 2.050E-05 | global batch size: 256 | lm loss: 3.321024E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.260 | TFLOPs: 30.29 | 7: iteration 20850/ 21553 | consumed samples: 5337600 | consumed tokens: 10931404800 | elapsed time per iteration (s): 0.30 | learning rate: 2.048E-05 | global batch size: 256 | lm loss: 3.325729E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.233 | TFLOPs: 30.25 | 7: iteration 20860/ 21553 | consumed samples: 5340160 | consumed tokens: 10936647680 | elapsed time per iteration (s): 0.30 | learning rate: 2.047E-05 | global batch size: 256 | lm loss: 3.324284E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.136 | TFLOPs: 30.25 | 7: iteration 20870/ 21553 | consumed samples: 5342720 | consumed tokens: 10941890560 | elapsed time per iteration (s): 0.30 | learning rate: 2.045E-05 | global batch size: 256 | lm loss: 3.330833E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.694 | TFLOPs: 30.24 | 7: iteration 20880/ 21553 | consumed samples: 5345280 | consumed tokens: 10947133440 | elapsed time per iteration (s): 0.30 | learning rate: 2.044E-05 | global batch size: 256 | lm loss: 3.315961E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.793 | TFLOPs: 30.24 | 7: iteration 20890/ 21553 | consumed samples: 5347840 | consumed tokens: 10952376320 | elapsed time per iteration (s): 0.30 | learning rate: 2.043E-05 | global batch size: 256 | lm loss: 3.325455E+00 | grad norm: 0.432 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 863.624 | TFLOPs: 30.23 | 7: iteration 20900/ 21553 | consumed samples: 5350400 | consumed tokens: 10957619200 | elapsed time per iteration (s): 0.30 | learning rate: 2.042E-05 | global batch size: 256 | lm loss: 3.324910E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.382 | TFLOPs: 30.26 | 7: iteration 20910/ 21553 | consumed samples: 5352960 | consumed tokens: 10962862080 | elapsed time per iteration (s): 0.30 | learning rate: 2.040E-05 | global batch size: 256 | lm loss: 3.324601E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.701 | TFLOPs: 30.27 | 7: iteration 20920/ 21553 | consumed samples: 5355520 | consumed tokens: 10968104960 | elapsed time per iteration (s): 0.30 | learning rate: 2.039E-05 | global batch size: 256 | lm loss: 3.323762E+00 | grad norm: 0.436 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.126 | TFLOPs: 30.29 | 7: iteration 20930/ 21553 | consumed samples: 5358080 | consumed tokens: 10973347840 | elapsed time per iteration (s): 0.30 | learning rate: 2.038E-05 | global batch size: 256 | lm loss: 3.325555E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.022 | TFLOPs: 30.28 | 7: iteration 20940/ 21553 | consumed samples: 5360640 | consumed tokens: 10978590720 | elapsed time per iteration (s): 0.30 | learning rate: 2.037E-05 | global batch size: 256 | lm loss: 3.324450E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.878 | TFLOPs: 30.28 | 7: iteration 20950/ 21553 | consumed samples: 5363200 | consumed tokens: 10983833600 | elapsed time per iteration (s): 0.30 | learning rate: 2.035E-05 | global batch size: 256 | lm loss: 3.325220E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.915 | TFLOPs: 30.28 | 7: iteration 20960/ 21553 | consumed samples: 5365760 | consumed tokens: 10989076480 | elapsed time per iteration (s): 0.30 | learning rate: 2.034E-05 | global batch size: 256 | lm loss: 3.328815E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.149 | TFLOPs: 30.29 | 7: iteration 20970/ 21553 | consumed samples: 5368320 | consumed tokens: 10994319360 | elapsed time per iteration (s): 0.30 | learning rate: 2.033E-05 | global batch size: 256 | lm loss: 3.318829E+00 | grad norm: 0.410 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.363 | TFLOPs: 30.29 | 7: iteration 20980/ 21553 | consumed samples: 5370880 | consumed tokens: 10999562240 | elapsed time per iteration (s): 0.30 | learning rate: 2.032E-05 | global batch size: 256 | lm loss: 3.327073E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.030 | TFLOPs: 30.28 | 7: iteration 20990/ 21553 | consumed samples: 5373440 | consumed tokens: 11004805120 | elapsed time per iteration (s): 0.30 | learning rate: 2.031E-05 | global batch size: 256 | lm loss: 3.326844E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.153 | TFLOPs: 30.29 | 7: iteration 21000/ 21553 | consumed samples: 5376000 | consumed tokens: 11010048000 | elapsed time per iteration (s): 0.30 | learning rate: 2.030E-05 | global batch size: 256 | lm loss: 3.321522E+00 | grad norm: 0.442 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.233 | TFLOPs: 30.29 | 7: ------------------------------------------------------------------------------------------------ 7: validation loss at iteration 21000 | lm loss value: 3.960697E+00 | lm loss PPL: 5.249391E+01 | 7: ------------------------------------------------------------------------------------------------ 0: saving checkpoint at iteration 21000 to checkpoints_146m14b100mdedup 0: [2023-03-17 01:39:58,538] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step21000 is begin to save! 0: [2023-03-17 01:39:58,542] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:39:58,627] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:39:58,628] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:39:58,643] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:39:58,643] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:39:58,658] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:39:58,658] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:39:58,673] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:39:58,673] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:39:58,687] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:39:58,688] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:39:58,702] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:39:58,702] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:39:58,717] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:39:58,717] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:39:58,732] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:39:58,732] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:39:58,746] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:39:58,747] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:39:58,761] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:39:58,761] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:39:58,776] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:39:58,776] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:39:58,791] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:39:58,791] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:39:58,806] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:39:58,806] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:39:58,821] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:39:58,821] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:39:58,835] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:39:58,836] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:39:58,850] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:39:58,850] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:39:58,851] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:39:58,852] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step21000/mp_rank_00_model_states.pt 0: [2023-03-17 01:39:58,852] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/mp_rank_00_model_states.pt... 0: [2023-03-17 01:39:58,854] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/mp_rank_00_model_states.pt. 0: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:39:58,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:39:58,904] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:39:58,905] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,905] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 2: [2023-03-17 01:39:58,905] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:39:58,905] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,905] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:39:58,908] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 3: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 3: [2023-03-17 01:39:58,908] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 3: [2023-03-17 01:39:58,908] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:39:58,908] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 01:39:58,908] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:39:58,908] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:39:58,909] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 5: [2023-03-17 01:39:58,908] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 0: [2023-03-17 01:39:58,909] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 5: [2023-03-17 01:39:58,909] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,909] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:39:58,909] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:39:58,909] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,909] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,909] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,909] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: [2023-03-17 01:39:58,909] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:39:58,909] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 0: [2023-03-17 01:39:58,909] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 5: [2023-03-17 01:39:58,910] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:39:58,910] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:39:58,910] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 01:39:58,910] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 01:39:58,910] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 5: [2023-03-17 01:39:58,910] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:39:58,910] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: [2023-03-17 01:39:58,910] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:39:58,910] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 0: [2023-03-17 01:39:58,910] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 5: [2023-03-17 01:39:58,910] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: [2023-03-17 01:39:58,910] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 3: [2023-03-17 01:39:58,910] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:39:58,910] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 7: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:39:58,911] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 7: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:39:58,911] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 6: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,911] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,911] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 01:39:58,911] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 6: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 6: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:39:58,912] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 4: [2023-03-17 01:39:58,912] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 7: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:39:58,912] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 4: [2023-03-17 01:39:58,912] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 5: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:39:58,912] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:39:58,912] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 6: [2023-03-17 01:39:58,913] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,913] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 01:39:58,913] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 5: [2023-03-17 01:39:58,913] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:39:58,913] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 01:39:58,913] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 7: [2023-03-17 01:39:58,913] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,913] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,913] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:39:58,913] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:39:58,913] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,913] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,913] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 01:39:58,913] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 1: [2023-03-17 01:39:58,914] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 6: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 6: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 2: [2023-03-17 01:39:58,911] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,911] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 2: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:39:58,912] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 2: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:39:58,912] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,912] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 2: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:39:58,914] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,914] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 7: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 2: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 7: [2023-03-17 01:39:58,914] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 01:39:58,914] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 7: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 2: [2023-03-17 01:39:58,914] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:39:58,914] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 4: [2023-03-17 01:39:58,914] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:39:58,914] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,915] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 5: [2023-03-17 01:39:58,915] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:39:58,915] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 01:39:58,915] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 4: [2023-03-17 01:39:58,915] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:39:58,915] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,915] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 3: [2023-03-17 01:39:58,916] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:39:58,916] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 01:39:58,916] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 3: [2023-03-17 01:39:58,916] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:39:58,916] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 3: [2023-03-17 01:39:58,916] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 4: [2023-03-17 01:39:58,917] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:39:58,917] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:39:58,917] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:39:58,917] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 7: [2023-03-17 01:39:58,917] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 1: [2023-03-17 01:39:58,917] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,917] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 7: [2023-03-17 01:39:58,917] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,917] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 7: [2023-03-17 01:39:58,917] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:39:58,917] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:39:58,917] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 5: [2023-03-17 01:39:58,918] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:39:58,918] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 01:39:58,918] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 4: [2023-03-17 01:39:58,919] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:39:58,919] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 01:39:58,919] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 6: [2023-03-17 01:39:58,919] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,919] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,919] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:39:58,919] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,919] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:39:58,919] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 6: [2023-03-17 01:39:58,919] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 2: [2023-03-17 01:39:58,919] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:39:58,919] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,922] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:39:58,922] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:39:58,922] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:39:58,922] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 1: [2023-03-17 01:39:58,922] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 01:39:58,922] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:39:58,922] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,922] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 1: [2023-03-17 01:39:58,922] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 3: [2023-03-17 01:39:58,924] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:39:58,924] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:39:58,924] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:39:58,924] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:39:58,924] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 01:39:58,924] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 3: [2023-03-17 01:39:58,924] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 3: [2023-03-17 01:39:58,924] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 3: [2023-03-17 01:39:58,924] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: [2023-03-17 01:39:58,934] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:39:58,934] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21000 is ready now! 0: successfully saved checkpoint at iteration 21000 to checkpoints_146m14b100mdedup 7: time (ms) | save-checkpoint: 404.05 7: iteration 21010/ 21553 | consumed samples: 5378560 | consumed tokens: 11015290880 | elapsed time per iteration (s): 0.35 | learning rate: 2.029E-05 | global batch size: 256 | lm loss: 3.320743E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 740.964 | TFLOPs: 25.94 | 7: iteration 21020/ 21553 | consumed samples: 5381120 | consumed tokens: 11020533760 | elapsed time per iteration (s): 0.30 | learning rate: 2.028E-05 | global batch size: 256 | lm loss: 3.326022E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.809 | TFLOPs: 30.31 | 7: iteration 21030/ 21553 | consumed samples: 5383680 | consumed tokens: 11025776640 | elapsed time per iteration (s): 0.30 | learning rate: 2.027E-05 | global batch size: 256 | lm loss: 3.322057E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.008 | TFLOPs: 30.32 | 7: iteration 21040/ 21553 | consumed samples: 5386240 | consumed tokens: 11031019520 | elapsed time per iteration (s): 0.30 | learning rate: 2.026E-05 | global batch size: 256 | lm loss: 3.323505E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.125 | TFLOPs: 30.29 | 7: iteration 21050/ 21553 | consumed samples: 5388800 | consumed tokens: 11036262400 | elapsed time per iteration (s): 0.30 | learning rate: 2.025E-05 | global batch size: 256 | lm loss: 3.320507E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.608 | TFLOPs: 30.27 | 7: iteration 21060/ 21553 | consumed samples: 5391360 | consumed tokens: 11041505280 | elapsed time per iteration (s): 0.30 | learning rate: 2.024E-05 | global batch size: 256 | lm loss: 3.317387E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.295 | TFLOPs: 30.29 | 7: iteration 21070/ 21553 | consumed samples: 5393920 | consumed tokens: 11046748160 | elapsed time per iteration (s): 0.30 | learning rate: 2.023E-05 | global batch size: 256 | lm loss: 3.323746E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.402 | TFLOPs: 30.30 | 7: iteration 21080/ 21553 | consumed samples: 5396480 | consumed tokens: 11051991040 | elapsed time per iteration (s): 0.30 | learning rate: 2.022E-05 | global batch size: 256 | lm loss: 3.322365E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.992 | TFLOPs: 30.28 | 7: iteration 21090/ 21553 | consumed samples: 5399040 | consumed tokens: 11057233920 | elapsed time per iteration (s): 0.30 | learning rate: 2.021E-05 | global batch size: 256 | lm loss: 3.319319E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.470 | TFLOPs: 30.30 | 7: iteration 21100/ 21553 | consumed samples: 5401600 | consumed tokens: 11062476800 | elapsed time per iteration (s): 0.30 | learning rate: 2.020E-05 | global batch size: 256 | lm loss: 3.321027E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.904 | TFLOPs: 30.28 | 7: iteration 21110/ 21553 | consumed samples: 5404160 | consumed tokens: 11067719680 | elapsed time per iteration (s): 0.30 | learning rate: 2.019E-05 | global batch size: 256 | lm loss: 3.320898E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.960 | TFLOPs: 30.28 | 7: iteration 21120/ 21553 | consumed samples: 5406720 | consumed tokens: 11072962560 | elapsed time per iteration (s): 0.30 | learning rate: 2.018E-05 | global batch size: 256 | lm loss: 3.316383E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.013 | TFLOPs: 30.28 | 7: iteration 21130/ 21553 | consumed samples: 5409280 | consumed tokens: 11078205440 | elapsed time per iteration (s): 0.30 | learning rate: 2.017E-05 | global batch size: 256 | lm loss: 3.322233E+00 | grad norm: 0.414 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.018 | TFLOPs: 30.28 | 7: iteration 21140/ 21553 | consumed samples: 5411840 | consumed tokens: 11083448320 | elapsed time per iteration (s): 0.30 | learning rate: 2.017E-05 | global batch size: 256 | lm loss: 3.326200E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.224 | TFLOPs: 30.29 | 7: iteration 21150/ 21553 | consumed samples: 5414400 | consumed tokens: 11088691200 | elapsed time per iteration (s): 0.30 | learning rate: 2.016E-05 | global batch size: 256 | lm loss: 3.325190E+00 | grad norm: 0.423 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.585 | TFLOPs: 30.30 | 7: iteration 21160/ 21553 | consumed samples: 5416960 | consumed tokens: 11093934080 | elapsed time per iteration (s): 0.30 | learning rate: 2.015E-05 | global batch size: 256 | lm loss: 3.320435E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.676 | TFLOPs: 30.30 | 7: iteration 21170/ 21553 | consumed samples: 5419520 | consumed tokens: 11099176960 | elapsed time per iteration (s): 0.30 | learning rate: 2.014E-05 | global batch size: 256 | lm loss: 3.325067E+00 | grad norm: 0.431 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.295 | TFLOPs: 30.29 | 7: iteration 21180/ 21553 | consumed samples: 5422080 | consumed tokens: 11104419840 | elapsed time per iteration (s): 0.30 | learning rate: 2.014E-05 | global batch size: 256 | lm loss: 3.322685E+00 | grad norm: 0.439 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.891 | TFLOPs: 30.31 | 7: iteration 21190/ 21553 | consumed samples: 5424640 | consumed tokens: 11109662720 | elapsed time per iteration (s): 0.30 | learning rate: 2.013E-05 | global batch size: 256 | lm loss: 3.327665E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.498 | TFLOPs: 30.30 | 7: iteration 21200/ 21553 | consumed samples: 5427200 | consumed tokens: 11114905600 | elapsed time per iteration (s): 0.30 | learning rate: 2.012E-05 | global batch size: 256 | lm loss: 3.320585E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.313 | TFLOPs: 30.29 | 7: iteration 21210/ 21553 | consumed samples: 5429760 | consumed tokens: 11120148480 | elapsed time per iteration (s): 0.30 | learning rate: 2.011E-05 | global batch size: 256 | lm loss: 3.323312E+00 | grad norm: 0.420 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.306 | TFLOPs: 30.29 | 7: iteration 21220/ 21553 | consumed samples: 5432320 | consumed tokens: 11125391360 | elapsed time per iteration (s): 0.30 | learning rate: 2.011E-05 | global batch size: 256 | lm loss: 3.324935E+00 | grad norm: 0.434 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.188 | TFLOPs: 30.29 | 7: iteration 21230/ 21553 | consumed samples: 5434880 | consumed tokens: 11130634240 | elapsed time per iteration (s): 0.30 | learning rate: 2.010E-05 | global batch size: 256 | lm loss: 3.323612E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.622 | TFLOPs: 30.30 | 7: iteration 21240/ 21553 | consumed samples: 5437440 | consumed tokens: 11135877120 | elapsed time per iteration (s): 0.30 | learning rate: 2.010E-05 | global batch size: 256 | lm loss: 3.325933E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.347 | TFLOPs: 30.29 | 7: iteration 21250/ 21553 | consumed samples: 5440000 | consumed tokens: 11141120000 | elapsed time per iteration (s): 0.30 | learning rate: 2.009E-05 | global batch size: 256 | lm loss: 3.321667E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.617 | TFLOPs: 30.30 | 7: iteration 21260/ 21553 | consumed samples: 5442560 | consumed tokens: 11146362880 | elapsed time per iteration (s): 0.30 | learning rate: 2.008E-05 | global batch size: 256 | lm loss: 3.321271E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.436 | TFLOPs: 30.30 | 7: iteration 21270/ 21553 | consumed samples: 5445120 | consumed tokens: 11151605760 | elapsed time per iteration (s): 0.30 | learning rate: 2.008E-05 | global batch size: 256 | lm loss: 3.325636E+00 | grad norm: 0.429 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.509 | TFLOPs: 30.30 | 7: iteration 21280/ 21553 | consumed samples: 5447680 | consumed tokens: 11156848640 | elapsed time per iteration (s): 0.30 | learning rate: 2.007E-05 | global batch size: 256 | lm loss: 3.325893E+00 | grad norm: 0.430 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.634 | TFLOPs: 30.30 | 7: iteration 21290/ 21553 | consumed samples: 5450240 | consumed tokens: 11162091520 | elapsed time per iteration (s): 0.30 | learning rate: 2.007E-05 | global batch size: 256 | lm loss: 3.317291E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.131 | TFLOPs: 30.29 | 7: iteration 21300/ 21553 | consumed samples: 5452800 | consumed tokens: 11167334400 | elapsed time per iteration (s): 0.30 | learning rate: 2.006E-05 | global batch size: 256 | lm loss: 3.324304E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.830 | TFLOPs: 30.31 | 7: iteration 21310/ 21553 | consumed samples: 5455360 | consumed tokens: 11172577280 | elapsed time per iteration (s): 0.30 | learning rate: 2.006E-05 | global batch size: 256 | lm loss: 3.316947E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.348 | TFLOPs: 30.26 | 7: iteration 21320/ 21553 | consumed samples: 5457920 | consumed tokens: 11177820160 | elapsed time per iteration (s): 0.30 | learning rate: 2.005E-05 | global batch size: 256 | lm loss: 3.321063E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.560 | TFLOPs: 30.27 | 7: iteration 21330/ 21553 | consumed samples: 5460480 | consumed tokens: 11183063040 | elapsed time per iteration (s): 0.30 | learning rate: 2.005E-05 | global batch size: 256 | lm loss: 3.323668E+00 | grad norm: 0.427 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.874 | TFLOPs: 30.28 | 7: iteration 21340/ 21553 | consumed samples: 5463040 | consumed tokens: 11188305920 | elapsed time per iteration (s): 0.30 | learning rate: 2.004E-05 | global batch size: 256 | lm loss: 3.318522E+00 | grad norm: 0.440 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.528 | TFLOPs: 30.30 | 7: iteration 21350/ 21553 | consumed samples: 5465600 | consumed tokens: 11193548800 | elapsed time per iteration (s): 0.30 | learning rate: 2.004E-05 | global batch size: 256 | lm loss: 3.327438E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.357 | TFLOPs: 30.29 | 7: iteration 21360/ 21553 | consumed samples: 5468160 | consumed tokens: 11198791680 | elapsed time per iteration (s): 0.30 | learning rate: 2.004E-05 | global batch size: 256 | lm loss: 3.322485E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.603 | TFLOPs: 30.30 | 7: iteration 21370/ 21553 | consumed samples: 5470720 | consumed tokens: 11204034560 | elapsed time per iteration (s): 0.30 | learning rate: 2.003E-05 | global batch size: 256 | lm loss: 3.315243E+00 | grad norm: 0.418 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.301 | TFLOPs: 30.29 | 7: iteration 21380/ 21553 | consumed samples: 5473280 | consumed tokens: 11209277440 | elapsed time per iteration (s): 0.30 | learning rate: 2.003E-05 | global batch size: 256 | lm loss: 3.323630E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.107 | TFLOPs: 30.28 | 7: iteration 21390/ 21553 | consumed samples: 5475840 | consumed tokens: 11214520320 | elapsed time per iteration (s): 0.30 | learning rate: 2.003E-05 | global batch size: 256 | lm loss: 3.321829E+00 | grad norm: 0.433 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.175 | TFLOPs: 30.29 | 7: iteration 21400/ 21553 | consumed samples: 5478400 | consumed tokens: 11219763200 | elapsed time per iteration (s): 0.30 | learning rate: 2.002E-05 | global batch size: 256 | lm loss: 3.318609E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.755 | TFLOPs: 30.31 | 7: iteration 21410/ 21553 | consumed samples: 5480960 | consumed tokens: 11225006080 | elapsed time per iteration (s): 0.30 | learning rate: 2.002E-05 | global batch size: 256 | lm loss: 3.327200E+00 | grad norm: 0.443 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.675 | TFLOPs: 30.30 | 7: iteration 21420/ 21553 | consumed samples: 5483520 | consumed tokens: 11230248960 | elapsed time per iteration (s): 0.30 | learning rate: 2.002E-05 | global batch size: 256 | lm loss: 3.321154E+00 | grad norm: 0.426 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.466 | TFLOPs: 30.30 | 7: iteration 21430/ 21553 | consumed samples: 5486080 | consumed tokens: 11235491840 | elapsed time per iteration (s): 0.30 | learning rate: 2.001E-05 | global batch size: 256 | lm loss: 3.317990E+00 | grad norm: 0.416 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.369 | TFLOPs: 30.29 | 7: iteration 21440/ 21553 | consumed samples: 5488640 | consumed tokens: 11240734720 | elapsed time per iteration (s): 0.30 | learning rate: 2.001E-05 | global batch size: 256 | lm loss: 3.321634E+00 | grad norm: 0.417 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.394 | TFLOPs: 30.30 | 7: iteration 21450/ 21553 | consumed samples: 5491200 | consumed tokens: 11245977600 | elapsed time per iteration (s): 0.30 | learning rate: 2.001E-05 | global batch size: 256 | lm loss: 3.326658E+00 | grad norm: 0.425 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 866.034 | TFLOPs: 30.32 | 7: iteration 21460/ 21553 | consumed samples: 5493760 | consumed tokens: 11251220480 | elapsed time per iteration (s): 0.30 | learning rate: 2.001E-05 | global batch size: 256 | lm loss: 3.324346E+00 | grad norm: 0.437 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.224 | TFLOPs: 30.29 | 7: iteration 21470/ 21553 | consumed samples: 5496320 | consumed tokens: 11256463360 | elapsed time per iteration (s): 0.30 | learning rate: 2.001E-05 | global batch size: 256 | lm loss: 3.320571E+00 | grad norm: 0.421 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.683 | TFLOPs: 30.31 | 7: iteration 21480/ 21553 | consumed samples: 5498880 | consumed tokens: 11261706240 | elapsed time per iteration (s): 0.30 | learning rate: 2.001E-05 | global batch size: 256 | lm loss: 3.323562E+00 | grad norm: 0.422 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.516 | TFLOPs: 30.30 | 7: iteration 21490/ 21553 | consumed samples: 5501440 | consumed tokens: 11266949120 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-05 | global batch size: 256 | lm loss: 3.327951E+00 | grad norm: 0.453 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.459 | TFLOPs: 30.30 | 7: iteration 21500/ 21553 | consumed samples: 5504000 | consumed tokens: 11272192000 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-05 | global batch size: 256 | lm loss: 3.324393E+00 | grad norm: 0.424 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.221 | TFLOPs: 30.29 | 7: iteration 21510/ 21553 | consumed samples: 5506560 | consumed tokens: 11277434880 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-05 | global batch size: 256 | lm loss: 3.313323E+00 | grad norm: 0.407 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.462 | TFLOPs: 30.30 | 7: iteration 21520/ 21553 | consumed samples: 5509120 | consumed tokens: 11282677760 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-05 | global batch size: 256 | lm loss: 3.313910E+00 | grad norm: 0.415 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.730 | TFLOPs: 30.27 | 7: iteration 21530/ 21553 | consumed samples: 5511680 | consumed tokens: 11287920640 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-05 | global batch size: 256 | lm loss: 3.312727E+00 | grad norm: 0.406 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.242 | TFLOPs: 30.29 | 7: iteration 21540/ 21553 | consumed samples: 5514240 | consumed tokens: 11293163520 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-05 | global batch size: 256 | lm loss: 3.315180E+00 | grad norm: 0.419 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 864.844 | TFLOPs: 30.28 | 7: iteration 21550/ 21553 | consumed samples: 5516800 | consumed tokens: 11298406400 | elapsed time per iteration (s): 0.30 | learning rate: 2.000E-05 | global batch size: 256 | lm loss: 3.318022E+00 | grad norm: 0.428 | num zeros: 0.0 | number of skipped iterations: 0 | number of nan iterations: 0 | samples per second: 865.711 | TFLOPs: 30.31 | 0: [after training is done] datetime: 2023-03-17 01:42:42 0: saving checkpoint at iteration 21553 to checkpoints_146m14b100mdedup 7: ----------------------------------------------------------------------------------------------------------------- 7: validation loss at the end of training for val data | lm loss value: 3.978714E+00 | lm loss PPL: 5.344824E+01 | 7: ----------------------------------------------------------------------------------------------------------------- 0: [2023-03-17 01:42:42,633] [INFO] [logging.py:68:log_dist] [Rank 0] [Torch] Checkpoint global_step21553 is begin to save! 0: [2023-03-17 01:42:42,637] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_01-model_00-model_states.pt... 0: [2023-03-17 01:42:42,722] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_01-model_00-model_states.pt. 0: [2023-03-17 01:42:42,722] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_03-model_00-model_states.pt... 0: [2023-03-17 01:42:42,738] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_03-model_00-model_states.pt. 0: [2023-03-17 01:42:42,739] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_04-model_00-model_states.pt... 0: [2023-03-17 01:42:42,753] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_04-model_00-model_states.pt. 0: [2023-03-17 01:42:42,754] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_05-model_00-model_states.pt... 0: [2023-03-17 01:42:42,768] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_05-model_00-model_states.pt. 0: [2023-03-17 01:42:42,769] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_06-model_00-model_states.pt... 0: [2023-03-17 01:42:42,783] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_06-model_00-model_states.pt. 0: [2023-03-17 01:42:42,783] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_07-model_00-model_states.pt... 0: [2023-03-17 01:42:42,798] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_07-model_00-model_states.pt. 0: [2023-03-17 01:42:42,798] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_08-model_00-model_states.pt... 0: [2023-03-17 01:42:42,813] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_08-model_00-model_states.pt. 0: [2023-03-17 01:42:42,813] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_09-model_00-model_states.pt... 0: [2023-03-17 01:42:42,828] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_09-model_00-model_states.pt. 0: [2023-03-17 01:42:42,828] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_10-model_00-model_states.pt... 0: [2023-03-17 01:42:42,842] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_10-model_00-model_states.pt. 0: [2023-03-17 01:42:42,843] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_11-model_00-model_states.pt... 0: [2023-03-17 01:42:42,857] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_11-model_00-model_states.pt. 0: [2023-03-17 01:42:42,857] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_12-model_00-model_states.pt... 0: [2023-03-17 01:42:42,872] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_12-model_00-model_states.pt. 0: [2023-03-17 01:42:42,872] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_13-model_00-model_states.pt... 0: [2023-03-17 01:42:42,887] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_13-model_00-model_states.pt. 0: [2023-03-17 01:42:42,887] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_14-model_00-model_states.pt... 0: [2023-03-17 01:42:42,902] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_14-model_00-model_states.pt. 0: [2023-03-17 01:42:42,902] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_15-model_00-model_states.pt... 0: [2023-03-17 01:42:42,917] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_15-model_00-model_states.pt. 0: [2023-03-17 01:42:42,917] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_16-model_00-model_states.pt... 0: [2023-03-17 01:42:42,933] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_16-model_00-model_states.pt. 0: [2023-03-17 01:42:42,934] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_17-model_00-model_states.pt... 0: [2023-03-17 01:42:42,948] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_17-model_00-model_states.pt. 0: [2023-03-17 01:42:42,948] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/layer_19-model_00-model_states.pt... 0: [2023-03-17 01:42:42,949] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/layer_19-model_00-model_states.pt. 0: [2023-03-17 01:42:42,950] [INFO] [logging.py:68:log_dist] [Rank 0] Saving model checkpoint: checkpoints_146m14b100mdedup/global_step21553/mp_rank_00_model_states.pt 0: [2023-03-17 01:42:42,950] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/mp_rank_00_model_states.pt... 0: [2023-03-17 01:42:42,952] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/mp_rank_00_model_states.pt. 0: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt... 4: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt... 3: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt... 5: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt... 2: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt... 1: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt... 6: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt... 7: [2023-03-17 01:42:42,969] [INFO] [torch_checkpoint_engine.py:15:save] [Torch] Saving checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt... 0: [2023-03-17 01:42:43,000] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:42:43,002] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:42:43,002] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt 2: [2023-03-17 01:42:43,002] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 0: [2023-03-17 01:42:43,004] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:42:43,004] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt 0: [2023-03-17 01:42:43,004] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 0: [2023-03-17 01:42:43,005] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:42:43,005] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt 0: [2023-03-17 01:42:43,005] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 3: [2023-03-17 01:42:43,005] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:42:43,005] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_24_mp_rank_00_optim_states.pt 3: [2023-03-17 01:42:43,005] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 5: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_46_mp_rank_00_optim_states.pt 3: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 5: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_26_mp_rank_00_optim_states.pt 3: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_25_mp_rank_00_optim_states.pt 3: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 5: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_40_mp_rank_00_optim_states.pt 5: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 3: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 3: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_29_mp_rank_00_optim_states.pt 3: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 5: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_42_mp_rank_00_optim_states.pt 5: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 0: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt 5: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 5: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_41_mp_rank_00_optim_states.pt 0: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 1: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt 0: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 1: [2023-03-17 01:42:43,006] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt 1: [2023-03-17 01:42:43,006] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 1: [2023-03-17 01:42:43,007] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:42:43,007] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt 1: [2023-03-17 01:42:43,007] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 6: [2023-03-17 01:42:43,007] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:42:43,007] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:42:43,007] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_55_mp_rank_00_optim_states.pt 6: [2023-03-17 01:42:43,007] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_50_mp_rank_00_optim_states.pt 6: [2023-03-17 01:42:43,007] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:42:43,007] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 6: [2023-03-17 01:42:43,007] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 6: [2023-03-17 01:42:43,007] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_49_mp_rank_00_optim_states.pt 6: [2023-03-17 01:42:43,007] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 2: [2023-03-17 01:42:43,007] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt 2: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 2: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt 2: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 7: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt 2: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt 2: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 2: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 1: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt 1: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 5: [2023-03-17 01:42:43,009] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:42:43,009] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_45_mp_rank_00_optim_states.pt 5: [2023-03-17 01:42:43,009] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 4: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_35_mp_rank_00_optim_states.pt 4: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_37_mp_rank_00_optim_states.pt 4: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_39_mp_rank_00_optim_states.pt 4: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_38_mp_rank_00_optim_states.pt 4: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_34_mp_rank_00_optim_states.pt 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_36_mp_rank_00_optim_states.pt 4: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 5: [2023-03-17 01:42:43,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_44_mp_rank_00_optim_states.pt 5: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 0: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:42:43,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt 0: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 2: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:42:43,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt 3: [2023-03-17 01:42:43,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_30_mp_rank_00_optim_states.pt 0: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 3: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 1: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:42:43,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt 1: [2023-03-17 01:42:43,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt 1: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 1: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 3: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:42:43,011] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_31_mp_rank_00_optim_states.pt 3: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 3: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:42:43,011] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_27_mp_rank_00_optim_states.pt 3: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 5: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:42:43,011] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_47_mp_rank_00_optim_states.pt 5: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 6: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:42:43,012] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt 1: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 0: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt. 0: [2023-03-17 01:42:43,012] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt 0: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 5: [2023-03-17 01:42:43,013] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt. 5: [2023-03-17 01:42:43,013] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_43_mp_rank_00_optim_states.pt 5: [2023-03-17 01:42:43,013] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 3: [2023-03-17 01:42:43,013] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:42:43,013] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt. 3: [2023-03-17 01:42:43,014] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_28_mp_rank_00_optim_states.pt 1: [2023-03-17 01:42:43,014] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt 3: [2023-03-17 01:42:43,014] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 1: [2023-03-17 01:42:43,014] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 2: [2023-03-17 01:42:43,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt 2: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 2: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:42:43,012] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt 2: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 2: [2023-03-17 01:42:43,014] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt. 2: [2023-03-17 01:42:43,014] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt 2: [2023-03-17 01:42:43,014] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 4: [2023-03-17 01:42:43,014] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:42:43,014] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt. 4: [2023-03-17 01:42:43,014] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_33_mp_rank_00_optim_states.pt 4: [2023-03-17 01:42:43,014] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_32_mp_rank_00_optim_states.pt 4: [2023-03-17 01:42:43,014] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 4: [2023-03-17 01:42:43,014] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 1: [2023-03-17 01:42:43,015] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt. 1: [2023-03-17 01:42:43,015] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt 1: [2023-03-17 01:42:43,015] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 0: [2023-03-17 01:42:43,016] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt 0: [2023-03-17 01:42:43,016] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 6: [2023-03-17 01:42:43,011] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_54_mp_rank_00_optim_states.pt 6: [2023-03-17 01:42:43,011] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_52_mp_rank_00_optim_states.pt 6: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 6: [2023-03-17 01:42:43,011] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 6: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:42:43,012] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_48_mp_rank_00_optim_states.pt 6: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 6: [2023-03-17 01:42:43,012] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_51_mp_rank_00_optim_states.pt 6: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 6: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt. 6: [2023-03-17 01:42:43,012] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_53_mp_rank_00_optim_states.pt 6: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 7: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_57_mp_rank_00_optim_states.pt 7: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 7: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:42:43,008] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_63_mp_rank_00_optim_states.pt 7: [2023-03-17 01:42:43,008] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 7: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:42:43,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_59_mp_rank_00_optim_states.pt 7: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 7: [2023-03-17 01:42:43,010] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_58_mp_rank_00_optim_states.pt 7: [2023-03-17 01:42:43,010] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 7: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:42:43,012] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_60_mp_rank_00_optim_states.pt 7: [2023-03-17 01:42:43,012] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 7: [2023-03-17 01:42:43,013] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:42:43,013] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_62_mp_rank_00_optim_states.pt 7: [2023-03-17 01:42:43,013] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 7: [2023-03-17 01:42:43,015] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:42:43,015] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_56_mp_rank_00_optim_states.pt 7: [2023-03-17 01:42:43,015] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 7: [2023-03-17 01:42:43,015] [INFO] [torch_checkpoint_engine.py:17:save] [Torch] Saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt. 7: [2023-03-17 01:42:43,015] [INFO] [engine.py:3213:_save_zero_checkpoint] bf16_zero checkpoint saved checkpoints_146m14b100mdedup/global_step21553/bf16_zero_pp_rank_61_mp_rank_00_optim_states.pt 7: [2023-03-17 01:42:43,015] [INFO] [torch_checkpoint_engine.py:27:commit] [Torch] Checkpoint global_step21553 is ready now! 0: successfully saved checkpoint at iteration 21553 to checkpoints_146m14b100mdedup END 3326961: Fri 17 Mar 2023 01:43:01 AM EET