Commit
·
0dfc0c1
1
Parent(s):
3329375
Upload folder using huggingface_hub
Browse files- README.md +21 -0
- adapter_config.json +21 -0
- adapter_model.bin +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +32 -0
- train_args.json +123 -0
- training_args.bin +3 -0
- training_logs.json +1067 -0
README.md
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- load_in_8bit: False
|
| 10 |
+
- load_in_4bit: True
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: nf4
|
| 16 |
+
- bnb_4bit_use_double_quant: True
|
| 17 |
+
- bnb_4bit_compute_dtype: float16
|
| 18 |
+
### Framework versions
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
- PEFT 0.5.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "models/llama-30b",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.0,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 64,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "SEQ_CLS"
|
| 21 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:431951954e408c8dff8b4432c9466a628b6cd4677ddbf704d0bbc5359267001f
|
| 3 |
+
size 409081617
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": true,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"__type": "AddedToken",
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": true,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
"clean_up_tokenization_spaces": false,
|
| 11 |
+
"eos_token": {
|
| 12 |
+
"__type": "AddedToken",
|
| 13 |
+
"content": "</s>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false
|
| 18 |
+
},
|
| 19 |
+
"model_max_length": 2048,
|
| 20 |
+
"pad_token": null,
|
| 21 |
+
"sp_model_kwargs": {},
|
| 22 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"__type": "AddedToken",
|
| 25 |
+
"content": "<unk>",
|
| 26 |
+
"lstrip": false,
|
| 27 |
+
"normalized": true,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
},
|
| 31 |
+
"use_default_system_prompt": true
|
| 32 |
+
}
|
train_args.json
ADDED
|
@@ -0,0 +1,123 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"initial_model_dir": "models/llama-30b",
|
| 3 |
+
"distribution_id": "math_textbook",
|
| 4 |
+
"date_trained": "11/10/2023 04:28:33",
|
| 5 |
+
"output_dir": "models/classify_lora/llama-30b-math_textbook",
|
| 6 |
+
"overwrite_output_dir": false,
|
| 7 |
+
"do_train": false,
|
| 8 |
+
"do_eval": true,
|
| 9 |
+
"do_predict": false,
|
| 10 |
+
"evaluation_strategy": "steps",
|
| 11 |
+
"prediction_loss_only": false,
|
| 12 |
+
"per_device_train_batch_size": 8,
|
| 13 |
+
"per_device_eval_batch_size": 8,
|
| 14 |
+
"per_gpu_train_batch_size": null,
|
| 15 |
+
"per_gpu_eval_batch_size": null,
|
| 16 |
+
"gradient_accumulation_steps": 1,
|
| 17 |
+
"eval_accumulation_steps": null,
|
| 18 |
+
"eval_delay": 0,
|
| 19 |
+
"learning_rate": 0.0002,
|
| 20 |
+
"weight_decay": 0.0,
|
| 21 |
+
"adam_beta1": 0.9,
|
| 22 |
+
"adam_beta2": 0.999,
|
| 23 |
+
"adam_epsilon": 1e-08,
|
| 24 |
+
"max_grad_norm": 0.3,
|
| 25 |
+
"num_train_epochs": 2,
|
| 26 |
+
"max_steps": 100,
|
| 27 |
+
"lr_scheduler_type": "constant",
|
| 28 |
+
"warmup_ratio": 0.03,
|
| 29 |
+
"warmup_steps": 0,
|
| 30 |
+
"log_level": "passive",
|
| 31 |
+
"log_level_replica": "warning",
|
| 32 |
+
"log_on_each_node": true,
|
| 33 |
+
"logging_dir": "models/classify_lora/llama-30b-math_textbook/runs/Oct11_02-16-48_compute-permanent-node-990",
|
| 34 |
+
"logging_strategy": "steps",
|
| 35 |
+
"logging_first_step": false,
|
| 36 |
+
"logging_steps": 1,
|
| 37 |
+
"logging_nan_inf_filter": true,
|
| 38 |
+
"save_strategy": "steps",
|
| 39 |
+
"save_steps": 25,
|
| 40 |
+
"save_total_limit": 0,
|
| 41 |
+
"save_safetensors": false,
|
| 42 |
+
"save_on_each_node": false,
|
| 43 |
+
"no_cuda": false,
|
| 44 |
+
"use_cpu": false,
|
| 45 |
+
"use_mps_device": false,
|
| 46 |
+
"seed": 42,
|
| 47 |
+
"data_seed": null,
|
| 48 |
+
"jit_mode_eval": false,
|
| 49 |
+
"use_ipex": false,
|
| 50 |
+
"bf16": false,
|
| 51 |
+
"fp16": false,
|
| 52 |
+
"fp16_opt_level": "O1",
|
| 53 |
+
"half_precision_backend": "auto",
|
| 54 |
+
"bf16_full_eval": false,
|
| 55 |
+
"fp16_full_eval": false,
|
| 56 |
+
"tf32": null,
|
| 57 |
+
"local_rank": 0,
|
| 58 |
+
"ddp_backend": null,
|
| 59 |
+
"tpu_num_cores": null,
|
| 60 |
+
"tpu_metrics_debug": false,
|
| 61 |
+
"debug": [],
|
| 62 |
+
"dataloader_drop_last": false,
|
| 63 |
+
"eval_steps": 25,
|
| 64 |
+
"dataloader_num_workers": 0,
|
| 65 |
+
"past_index": -1,
|
| 66 |
+
"run_name": "train|models-classify_lora-llama-30b-math_textbook",
|
| 67 |
+
"disable_tqdm": false,
|
| 68 |
+
"remove_unused_columns": false,
|
| 69 |
+
"label_names": null,
|
| 70 |
+
"load_best_model_at_end": false,
|
| 71 |
+
"metric_for_best_model": "eval_math_textbook_score",
|
| 72 |
+
"greater_is_better": true,
|
| 73 |
+
"ignore_data_skip": false,
|
| 74 |
+
"sharded_ddp": [],
|
| 75 |
+
"fsdp": [],
|
| 76 |
+
"fsdp_min_num_params": 0,
|
| 77 |
+
"fsdp_config": {
|
| 78 |
+
"min_num_params": 0,
|
| 79 |
+
"xla": false,
|
| 80 |
+
"xla_fsdp_grad_ckpt": false
|
| 81 |
+
},
|
| 82 |
+
"fsdp_transformer_layer_cls_to_wrap": null,
|
| 83 |
+
"deepspeed": "configs/ds_zero_1.json",
|
| 84 |
+
"label_smoothing_factor": 0.0,
|
| 85 |
+
"optim": "paged_adamw_32bit",
|
| 86 |
+
"optim_args": null,
|
| 87 |
+
"adafactor": false,
|
| 88 |
+
"group_by_length": false,
|
| 89 |
+
"length_column_name": "length",
|
| 90 |
+
"report_to": [
|
| 91 |
+
"wandb"
|
| 92 |
+
],
|
| 93 |
+
"ddp_find_unused_parameters": false,
|
| 94 |
+
"ddp_bucket_cap_mb": null,
|
| 95 |
+
"ddp_broadcast_buffers": null,
|
| 96 |
+
"dataloader_pin_memory": true,
|
| 97 |
+
"skip_memory_metrics": true,
|
| 98 |
+
"use_legacy_prediction_loop": false,
|
| 99 |
+
"push_to_hub": false,
|
| 100 |
+
"resume_from_checkpoint": null,
|
| 101 |
+
"hub_model_id": null,
|
| 102 |
+
"hub_strategy": "every_save",
|
| 103 |
+
"hub_token": null,
|
| 104 |
+
"hub_private_repo": false,
|
| 105 |
+
"hub_always_push": false,
|
| 106 |
+
"gradient_checkpointing": false,
|
| 107 |
+
"include_inputs_for_metrics": false,
|
| 108 |
+
"fp16_backend": "auto",
|
| 109 |
+
"push_to_hub_model_id": null,
|
| 110 |
+
"push_to_hub_organization": null,
|
| 111 |
+
"push_to_hub_token": null,
|
| 112 |
+
"_n_gpu": 1,
|
| 113 |
+
"mp_parameters": "",
|
| 114 |
+
"auto_find_batch_size": false,
|
| 115 |
+
"full_determinism": false,
|
| 116 |
+
"torchdynamo": null,
|
| 117 |
+
"ray_scope": "last",
|
| 118 |
+
"ddp_timeout": 1800,
|
| 119 |
+
"torch_compile": false,
|
| 120 |
+
"torch_compile_backend": null,
|
| 121 |
+
"torch_compile_mode": null,
|
| 122 |
+
"dispatch_batches": null
|
| 123 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51ecfc818ea40241f2fb6f9e5e88adfc88ef05f40c4b28d289d40ffb82cff3b5
|
| 3 |
+
size 5115
|
training_logs.json
ADDED
|
@@ -0,0 +1,1067 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"loss": 0.7607,
|
| 4 |
+
"learning_rate": 0.0002,
|
| 5 |
+
"epoch": 0.02,
|
| 6 |
+
"step": 1
|
| 7 |
+
},
|
| 8 |
+
{
|
| 9 |
+
"loss": 0.7232,
|
| 10 |
+
"learning_rate": 0.0002,
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"step": 2
|
| 13 |
+
},
|
| 14 |
+
{
|
| 15 |
+
"loss": 0.6848,
|
| 16 |
+
"learning_rate": 0.0002,
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"step": 3
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"loss": 0.7019,
|
| 22 |
+
"learning_rate": 0.0002,
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"step": 4
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"loss": 0.6913,
|
| 28 |
+
"learning_rate": 0.0002,
|
| 29 |
+
"epoch": 0.09,
|
| 30 |
+
"step": 5
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"loss": 0.7081,
|
| 34 |
+
"learning_rate": 0.0002,
|
| 35 |
+
"epoch": 0.11,
|
| 36 |
+
"step": 6
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"loss": 0.6857,
|
| 40 |
+
"learning_rate": 0.0002,
|
| 41 |
+
"epoch": 0.12,
|
| 42 |
+
"step": 7
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"loss": 0.6294,
|
| 46 |
+
"learning_rate": 0.0002,
|
| 47 |
+
"epoch": 0.14,
|
| 48 |
+
"step": 8
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"loss": 0.6635,
|
| 52 |
+
"learning_rate": 0.0002,
|
| 53 |
+
"epoch": 0.16,
|
| 54 |
+
"step": 9
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"loss": 0.7302,
|
| 58 |
+
"learning_rate": 0.0002,
|
| 59 |
+
"epoch": 0.18,
|
| 60 |
+
"step": 10
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"loss": 0.6197,
|
| 64 |
+
"learning_rate": 0.0002,
|
| 65 |
+
"epoch": 0.19,
|
| 66 |
+
"step": 11
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"loss": 0.572,
|
| 70 |
+
"learning_rate": 0.0002,
|
| 71 |
+
"epoch": 0.21,
|
| 72 |
+
"step": 12
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"loss": 0.6062,
|
| 76 |
+
"learning_rate": 0.0002,
|
| 77 |
+
"epoch": 0.23,
|
| 78 |
+
"step": 13
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"loss": 0.5423,
|
| 82 |
+
"learning_rate": 0.0002,
|
| 83 |
+
"epoch": 0.25,
|
| 84 |
+
"step": 14
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"loss": 0.4641,
|
| 88 |
+
"learning_rate": 0.0002,
|
| 89 |
+
"epoch": 0.26,
|
| 90 |
+
"step": 15
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"loss": 0.4541,
|
| 94 |
+
"learning_rate": 0.0002,
|
| 95 |
+
"epoch": 0.28,
|
| 96 |
+
"step": 16
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"loss": 0.3941,
|
| 100 |
+
"learning_rate": 0.0002,
|
| 101 |
+
"epoch": 0.3,
|
| 102 |
+
"step": 17
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"loss": 0.4322,
|
| 106 |
+
"learning_rate": 0.0002,
|
| 107 |
+
"epoch": 0.32,
|
| 108 |
+
"step": 18
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"loss": 0.3091,
|
| 112 |
+
"learning_rate": 0.0002,
|
| 113 |
+
"epoch": 0.33,
|
| 114 |
+
"step": 19
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"loss": 0.3358,
|
| 118 |
+
"learning_rate": 0.0002,
|
| 119 |
+
"epoch": 0.35,
|
| 120 |
+
"step": 20
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"loss": 0.5353,
|
| 124 |
+
"learning_rate": 0.0002,
|
| 125 |
+
"epoch": 0.37,
|
| 126 |
+
"step": 21
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"loss": 0.2493,
|
| 130 |
+
"learning_rate": 0.0002,
|
| 131 |
+
"epoch": 0.39,
|
| 132 |
+
"step": 22
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"loss": 0.3774,
|
| 136 |
+
"learning_rate": 0.0002,
|
| 137 |
+
"epoch": 0.4,
|
| 138 |
+
"step": 23
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"loss": 0.4386,
|
| 142 |
+
"learning_rate": 0.0002,
|
| 143 |
+
"epoch": 0.42,
|
| 144 |
+
"step": 24
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"loss": 0.2575,
|
| 148 |
+
"learning_rate": 0.0002,
|
| 149 |
+
"epoch": 0.44,
|
| 150 |
+
"step": 25
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"eval_math_textbook_loss": 0.16202543675899506,
|
| 154 |
+
"eval_math_textbook_score": -0.05223832651972771,
|
| 155 |
+
"eval_math_textbook_brier_score": 0.05223832651972771,
|
| 156 |
+
"eval_math_textbook_average_probability": 0.8756378889083862,
|
| 157 |
+
"eval_math_textbook_accuracy": 0.94,
|
| 158 |
+
"eval_math_textbook_probabilities": [
|
| 159 |
+
0.9960231781005859,
|
| 160 |
+
0.9589232802391052,
|
| 161 |
+
0.9999951124191284,
|
| 162 |
+
0.9927495718002319,
|
| 163 |
+
0.9983804225921631,
|
| 164 |
+
0.9993534684181213,
|
| 165 |
+
0.49611395597457886,
|
| 166 |
+
0.6533418893814087,
|
| 167 |
+
0.7788395881652832,
|
| 168 |
+
0.6563414931297302,
|
| 169 |
+
0.2655615508556366,
|
| 170 |
+
0.5199369192123413,
|
| 171 |
+
0.9290330410003662,
|
| 172 |
+
0.9862822890281677,
|
| 173 |
+
0.9882707595825195,
|
| 174 |
+
0.9478281736373901,
|
| 175 |
+
0.9970117807388306,
|
| 176 |
+
0.9931566119194031,
|
| 177 |
+
0.9889145493507385,
|
| 178 |
+
0.9937031865119934,
|
| 179 |
+
0.9999682903289795,
|
| 180 |
+
0.7295100092887878,
|
| 181 |
+
0.9783802628517151,
|
| 182 |
+
0.9998266100883484,
|
| 183 |
+
0.9879601001739502,
|
| 184 |
+
0.9995874762535095,
|
| 185 |
+
0.9808893799781799,
|
| 186 |
+
0.8457397818565369,
|
| 187 |
+
0.9344746470451355,
|
| 188 |
+
0.876669704914093,
|
| 189 |
+
0.1977093666791916,
|
| 190 |
+
0.9969134330749512,
|
| 191 |
+
0.9943760633468628,
|
| 192 |
+
0.9675971865653992,
|
| 193 |
+
0.9927055239677429,
|
| 194 |
+
0.9609069228172302,
|
| 195 |
+
0.41270166635513306,
|
| 196 |
+
0.5432194471359253,
|
| 197 |
+
0.7506809234619141,
|
| 198 |
+
0.8879430294036865,
|
| 199 |
+
0.9209844470024109,
|
| 200 |
+
0.9765685796737671,
|
| 201 |
+
0.986634373664856,
|
| 202 |
+
0.9965680837631226,
|
| 203 |
+
0.9989145994186401,
|
| 204 |
+
0.9733389616012573,
|
| 205 |
+
0.9279597997665405,
|
| 206 |
+
0.926964521408081,
|
| 207 |
+
0.9984092116355896,
|
| 208 |
+
0.993224561214447,
|
| 209 |
+
0.9959009289741516,
|
| 210 |
+
0.9766042828559875,
|
| 211 |
+
0.99498051404953,
|
| 212 |
+
0.9953020811080933,
|
| 213 |
+
0.7484961748123169,
|
| 214 |
+
0.5595658421516418,
|
| 215 |
+
0.6347947120666504,
|
| 216 |
+
0.8333064913749695,
|
| 217 |
+
0.8934006094932556,
|
| 218 |
+
0.8698863387107849,
|
| 219 |
+
0.9989472031593323,
|
| 220 |
+
0.9997424483299255,
|
| 221 |
+
0.9983856678009033,
|
| 222 |
+
0.5097801089286804,
|
| 223 |
+
0.8684792518615723,
|
| 224 |
+
0.861236572265625,
|
| 225 |
+
0.6593285799026489,
|
| 226 |
+
0.9650273323059082,
|
| 227 |
+
0.9815667867660522,
|
| 228 |
+
0.8728976845741272,
|
| 229 |
+
0.9385504126548767,
|
| 230 |
+
0.8316371440887451,
|
| 231 |
+
0.9991797804832458,
|
| 232 |
+
0.9993802309036255,
|
| 233 |
+
0.999068558216095,
|
| 234 |
+
0.9983657002449036,
|
| 235 |
+
0.9968892931938171,
|
| 236 |
+
0.989550769329071,
|
| 237 |
+
0.9781104922294617,
|
| 238 |
+
0.9844730496406555,
|
| 239 |
+
0.6861022114753723,
|
| 240 |
+
0.9963429570198059,
|
| 241 |
+
0.9961416125297546,
|
| 242 |
+
0.9998307228088379,
|
| 243 |
+
0.9972301125526428,
|
| 244 |
+
0.9972461462020874,
|
| 245 |
+
0.9998589754104614,
|
| 246 |
+
0.8539162874221802,
|
| 247 |
+
0.8533269166946411,
|
| 248 |
+
0.9063481092453003,
|
| 249 |
+
0.9272497892379761,
|
| 250 |
+
0.2949357330799103,
|
| 251 |
+
0.9830122590065002,
|
| 252 |
+
0.9900804162025452,
|
| 253 |
+
0.9989221096038818,
|
| 254 |
+
0.9978867173194885,
|
| 255 |
+
0.7507686018943787,
|
| 256 |
+
0.2280157506465912,
|
| 257 |
+
0.5014479160308838,
|
| 258 |
+
0.7952038049697876
|
| 259 |
+
],
|
| 260 |
+
"eval_math_textbook_runtime": 88.1152,
|
| 261 |
+
"eval_math_textbook_samples_per_second": 1.135,
|
| 262 |
+
"eval_math_textbook_steps_per_second": 0.045,
|
| 263 |
+
"epoch": 0.44,
|
| 264 |
+
"step": 25
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"loss": 0.06,
|
| 268 |
+
"learning_rate": 0.0002,
|
| 269 |
+
"epoch": 0.46,
|
| 270 |
+
"step": 26
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"loss": 0.2526,
|
| 274 |
+
"learning_rate": 0.0002,
|
| 275 |
+
"epoch": 0.47,
|
| 276 |
+
"step": 27
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"loss": 0.1444,
|
| 280 |
+
"learning_rate": 0.0002,
|
| 281 |
+
"epoch": 0.49,
|
| 282 |
+
"step": 28
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"loss": 0.1755,
|
| 286 |
+
"learning_rate": 0.0002,
|
| 287 |
+
"epoch": 0.51,
|
| 288 |
+
"step": 29
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"loss": 0.0442,
|
| 292 |
+
"learning_rate": 0.0002,
|
| 293 |
+
"epoch": 0.53,
|
| 294 |
+
"step": 30
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"loss": 0.3503,
|
| 298 |
+
"learning_rate": 0.0002,
|
| 299 |
+
"epoch": 0.54,
|
| 300 |
+
"step": 31
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"loss": 0.469,
|
| 304 |
+
"learning_rate": 0.0002,
|
| 305 |
+
"epoch": 0.56,
|
| 306 |
+
"step": 32
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"loss": 0.135,
|
| 310 |
+
"learning_rate": 0.0002,
|
| 311 |
+
"epoch": 0.58,
|
| 312 |
+
"step": 33
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"loss": 0.1248,
|
| 316 |
+
"learning_rate": 0.0002,
|
| 317 |
+
"epoch": 0.6,
|
| 318 |
+
"step": 34
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"loss": 0.1386,
|
| 322 |
+
"learning_rate": 0.0002,
|
| 323 |
+
"epoch": 0.61,
|
| 324 |
+
"step": 35
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"loss": 0.1263,
|
| 328 |
+
"learning_rate": 0.0002,
|
| 329 |
+
"epoch": 0.63,
|
| 330 |
+
"step": 36
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"loss": 0.0569,
|
| 334 |
+
"learning_rate": 0.0002,
|
| 335 |
+
"epoch": 0.65,
|
| 336 |
+
"step": 37
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"loss": 0.0416,
|
| 340 |
+
"learning_rate": 0.0002,
|
| 341 |
+
"epoch": 0.67,
|
| 342 |
+
"step": 38
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"loss": 0.0964,
|
| 346 |
+
"learning_rate": 0.0002,
|
| 347 |
+
"epoch": 0.68,
|
| 348 |
+
"step": 39
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"loss": 0.1388,
|
| 352 |
+
"learning_rate": 0.0002,
|
| 353 |
+
"epoch": 0.7,
|
| 354 |
+
"step": 40
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"loss": 0.3447,
|
| 358 |
+
"learning_rate": 0.0002,
|
| 359 |
+
"epoch": 0.72,
|
| 360 |
+
"step": 41
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"loss": 0.0337,
|
| 364 |
+
"learning_rate": 0.0002,
|
| 365 |
+
"epoch": 0.74,
|
| 366 |
+
"step": 42
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"loss": 0.2434,
|
| 370 |
+
"learning_rate": 0.0002,
|
| 371 |
+
"epoch": 0.75,
|
| 372 |
+
"step": 43
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"loss": 0.0229,
|
| 376 |
+
"learning_rate": 0.0002,
|
| 377 |
+
"epoch": 0.77,
|
| 378 |
+
"step": 44
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"loss": 0.1047,
|
| 382 |
+
"learning_rate": 0.0002,
|
| 383 |
+
"epoch": 0.79,
|
| 384 |
+
"step": 45
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"loss": 0.1124,
|
| 388 |
+
"learning_rate": 0.0002,
|
| 389 |
+
"epoch": 0.81,
|
| 390 |
+
"step": 46
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"loss": 0.2248,
|
| 394 |
+
"learning_rate": 0.0002,
|
| 395 |
+
"epoch": 0.82,
|
| 396 |
+
"step": 47
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"loss": 0.1916,
|
| 400 |
+
"learning_rate": 0.0002,
|
| 401 |
+
"epoch": 0.84,
|
| 402 |
+
"step": 48
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"loss": 0.0577,
|
| 406 |
+
"learning_rate": 0.0002,
|
| 407 |
+
"epoch": 0.86,
|
| 408 |
+
"step": 49
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"loss": 0.0294,
|
| 412 |
+
"learning_rate": 0.0002,
|
| 413 |
+
"epoch": 0.88,
|
| 414 |
+
"step": 50
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"eval_math_textbook_loss": 0.062490738928318024,
|
| 418 |
+
"eval_math_textbook_score": -0.02642030455172062,
|
| 419 |
+
"eval_math_textbook_brier_score": 0.02642030455172062,
|
| 420 |
+
"eval_math_textbook_average_probability": 0.9505031108856201,
|
| 421 |
+
"eval_math_textbook_accuracy": 0.96,
|
| 422 |
+
"eval_math_textbook_probabilities": [
|
| 423 |
+
0.9935856461524963,
|
| 424 |
+
0.9854414463043213,
|
| 425 |
+
0.9999929666519165,
|
| 426 |
+
0.9998990297317505,
|
| 427 |
+
0.9999942779541016,
|
| 428 |
+
0.9999994039535522,
|
| 429 |
+
0.8846103549003601,
|
| 430 |
+
0.9717264175415039,
|
| 431 |
+
0.9865896701812744,
|
| 432 |
+
0.9265250563621521,
|
| 433 |
+
0.7362580299377441,
|
| 434 |
+
0.9999997615814209,
|
| 435 |
+
0.9999853372573853,
|
| 436 |
+
0.9999990463256836,
|
| 437 |
+
0.9999983310699463,
|
| 438 |
+
0.8599734306335449,
|
| 439 |
+
0.9999970197677612,
|
| 440 |
+
0.9999957084655762,
|
| 441 |
+
0.9904043674468994,
|
| 442 |
+
0.990830659866333,
|
| 443 |
+
0.9999998807907104,
|
| 444 |
+
0.8436688780784607,
|
| 445 |
+
0.9999500513076782,
|
| 446 |
+
0.9999864101409912,
|
| 447 |
+
0.9999352693557739,
|
| 448 |
+
1.0,
|
| 449 |
+
0.9993420243263245,
|
| 450 |
+
0.9999982118606567,
|
| 451 |
+
0.9998797178268433,
|
| 452 |
+
0.9999988079071045,
|
| 453 |
+
0.9999676942825317,
|
| 454 |
+
0.9999847412109375,
|
| 455 |
+
0.999996542930603,
|
| 456 |
+
0.9981550574302673,
|
| 457 |
+
0.9999818801879883,
|
| 458 |
+
0.999930739402771,
|
| 459 |
+
0.8088546395301819,
|
| 460 |
+
0.9683941602706909,
|
| 461 |
+
0.9999778270721436,
|
| 462 |
+
0.9500603079795837,
|
| 463 |
+
0.7374384999275208,
|
| 464 |
+
0.9987793564796448,
|
| 465 |
+
0.9999827146530151,
|
| 466 |
+
0.9997771382331848,
|
| 467 |
+
0.9933000802993774,
|
| 468 |
+
0.9999254941940308,
|
| 469 |
+
0.9691327810287476,
|
| 470 |
+
0.9999750852584839,
|
| 471 |
+
1.0,
|
| 472 |
+
1.0,
|
| 473 |
+
1.0,
|
| 474 |
+
0.9915546774864197,
|
| 475 |
+
0.9993717074394226,
|
| 476 |
+
0.9983396530151367,
|
| 477 |
+
0.8835324645042419,
|
| 478 |
+
0.6976872682571411,
|
| 479 |
+
0.9402629137039185,
|
| 480 |
+
0.9999927282333374,
|
| 481 |
+
0.999998927116394,
|
| 482 |
+
0.9999991655349731,
|
| 483 |
+
0.9999998807907104,
|
| 484 |
+
1.0,
|
| 485 |
+
0.9999986886978149,
|
| 486 |
+
0.9999734163284302,
|
| 487 |
+
0.9999517202377319,
|
| 488 |
+
0.9999568462371826,
|
| 489 |
+
0.9999176263809204,
|
| 490 |
+
0.9999970197677612,
|
| 491 |
+
0.9999969005584717,
|
| 492 |
+
0.9998503923416138,
|
| 493 |
+
0.9999924898147583,
|
| 494 |
+
0.9999098777770996,
|
| 495 |
+
0.9999880790710449,
|
| 496 |
+
0.9999984502792358,
|
| 497 |
+
0.9999986886978149,
|
| 498 |
+
0.9852681159973145,
|
| 499 |
+
0.9834801554679871,
|
| 500 |
+
0.9828230738639832,
|
| 501 |
+
0.999982476234436,
|
| 502 |
+
0.999993085861206,
|
| 503 |
+
0.995590090751648,
|
| 504 |
+
1.0,
|
| 505 |
+
0.9999998807907104,
|
| 506 |
+
1.0,
|
| 507 |
+
0.99729984998703,
|
| 508 |
+
0.9949920773506165,
|
| 509 |
+
0.9999861717224121,
|
| 510 |
+
0.9991029500961304,
|
| 511 |
+
0.9995905756950378,
|
| 512 |
+
0.9950645565986633,
|
| 513 |
+
0.9799520373344421,
|
| 514 |
+
0.43311816453933716,
|
| 515 |
+
0.9996745586395264,
|
| 516 |
+
0.999977707862854,
|
| 517 |
+
0.9999972581863403,
|
| 518 |
+
0.9999926090240479,
|
| 519 |
+
0.3268202543258667,
|
| 520 |
+
0.02255621738731861,
|
| 521 |
+
0.25264206528663635,
|
| 522 |
+
0.9999605417251587
|
| 523 |
+
],
|
| 524 |
+
"eval_math_textbook_runtime": 88.0991,
|
| 525 |
+
"eval_math_textbook_samples_per_second": 1.135,
|
| 526 |
+
"eval_math_textbook_steps_per_second": 0.045,
|
| 527 |
+
"epoch": 0.88,
|
| 528 |
+
"step": 50
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"loss": 0.0318,
|
| 532 |
+
"learning_rate": 0.0002,
|
| 533 |
+
"epoch": 0.89,
|
| 534 |
+
"step": 51
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"loss": 0.0211,
|
| 538 |
+
"learning_rate": 0.0002,
|
| 539 |
+
"epoch": 0.91,
|
| 540 |
+
"step": 52
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"loss": 0.0557,
|
| 544 |
+
"learning_rate": 0.0002,
|
| 545 |
+
"epoch": 0.93,
|
| 546 |
+
"step": 53
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"loss": 0.0605,
|
| 550 |
+
"learning_rate": 0.0002,
|
| 551 |
+
"epoch": 0.95,
|
| 552 |
+
"step": 54
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"loss": 0.1855,
|
| 556 |
+
"learning_rate": 0.0002,
|
| 557 |
+
"epoch": 0.96,
|
| 558 |
+
"step": 55
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"loss": 0.0538,
|
| 562 |
+
"learning_rate": 0.0002,
|
| 563 |
+
"epoch": 0.98,
|
| 564 |
+
"step": 56
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"loss": 0.0276,
|
| 568 |
+
"learning_rate": 0.0002,
|
| 569 |
+
"epoch": 1.0,
|
| 570 |
+
"step": 57
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"loss": 0.0116,
|
| 574 |
+
"learning_rate": 0.0002,
|
| 575 |
+
"epoch": 1.02,
|
| 576 |
+
"step": 58
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"loss": 0.0469,
|
| 580 |
+
"learning_rate": 0.0002,
|
| 581 |
+
"epoch": 1.04,
|
| 582 |
+
"step": 59
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"loss": 0.1541,
|
| 586 |
+
"learning_rate": 0.0002,
|
| 587 |
+
"epoch": 1.05,
|
| 588 |
+
"step": 60
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"loss": 0.0957,
|
| 592 |
+
"learning_rate": 0.0002,
|
| 593 |
+
"epoch": 1.07,
|
| 594 |
+
"step": 61
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"loss": 0.0305,
|
| 598 |
+
"learning_rate": 0.0002,
|
| 599 |
+
"epoch": 1.09,
|
| 600 |
+
"step": 62
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"loss": 0.0596,
|
| 604 |
+
"learning_rate": 0.0002,
|
| 605 |
+
"epoch": 1.11,
|
| 606 |
+
"step": 63
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"loss": 0.0154,
|
| 610 |
+
"learning_rate": 0.0002,
|
| 611 |
+
"epoch": 1.12,
|
| 612 |
+
"step": 64
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"loss": 0.0505,
|
| 616 |
+
"learning_rate": 0.0002,
|
| 617 |
+
"epoch": 1.14,
|
| 618 |
+
"step": 65
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"loss": 0.1137,
|
| 622 |
+
"learning_rate": 0.0002,
|
| 623 |
+
"epoch": 1.16,
|
| 624 |
+
"step": 66
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"loss": 0.0445,
|
| 628 |
+
"learning_rate": 0.0002,
|
| 629 |
+
"epoch": 1.18,
|
| 630 |
+
"step": 67
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"loss": 0.0243,
|
| 634 |
+
"learning_rate": 0.0002,
|
| 635 |
+
"epoch": 1.19,
|
| 636 |
+
"step": 68
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"loss": 0.0427,
|
| 640 |
+
"learning_rate": 0.0002,
|
| 641 |
+
"epoch": 1.21,
|
| 642 |
+
"step": 69
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"loss": 0.0269,
|
| 646 |
+
"learning_rate": 0.0002,
|
| 647 |
+
"epoch": 1.23,
|
| 648 |
+
"step": 70
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"loss": 0.0033,
|
| 652 |
+
"learning_rate": 0.0002,
|
| 653 |
+
"epoch": 1.25,
|
| 654 |
+
"step": 71
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"loss": 0.0753,
|
| 658 |
+
"learning_rate": 0.0002,
|
| 659 |
+
"epoch": 1.26,
|
| 660 |
+
"step": 72
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"loss": 0.0136,
|
| 664 |
+
"learning_rate": 0.0002,
|
| 665 |
+
"epoch": 1.28,
|
| 666 |
+
"step": 73
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"loss": 0.0148,
|
| 670 |
+
"learning_rate": 0.0002,
|
| 671 |
+
"epoch": 1.3,
|
| 672 |
+
"step": 74
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"loss": 0.0294,
|
| 676 |
+
"learning_rate": 0.0002,
|
| 677 |
+
"epoch": 1.32,
|
| 678 |
+
"step": 75
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"eval_math_textbook_loss": 0.07301133126020432,
|
| 682 |
+
"eval_math_textbook_score": -0.03377831354737282,
|
| 683 |
+
"eval_math_textbook_brier_score": 0.03377831354737282,
|
| 684 |
+
"eval_math_textbook_average_probability": 0.9545721411705017,
|
| 685 |
+
"eval_math_textbook_accuracy": 0.95,
|
| 686 |
+
"eval_math_textbook_probabilities": [
|
| 687 |
+
0.9997319579124451,
|
| 688 |
+
0.998867392539978,
|
| 689 |
+
0.9999995231628418,
|
| 690 |
+
1.0,
|
| 691 |
+
1.0,
|
| 692 |
+
1.0,
|
| 693 |
+
0.9868295788764954,
|
| 694 |
+
0.999461829662323,
|
| 695 |
+
0.9999980926513672,
|
| 696 |
+
0.9989113807678223,
|
| 697 |
+
0.8771836161613464,
|
| 698 |
+
1.0,
|
| 699 |
+
1.0,
|
| 700 |
+
1.0,
|
| 701 |
+
1.0,
|
| 702 |
+
0.9248412847518921,
|
| 703 |
+
0.9999861717224121,
|
| 704 |
+
0.9999493360519409,
|
| 705 |
+
0.9935328960418701,
|
| 706 |
+
0.9922099113464355,
|
| 707 |
+
1.0,
|
| 708 |
+
0.9558141231536865,
|
| 709 |
+
0.9999998807907104,
|
| 710 |
+
1.0,
|
| 711 |
+
1.0,
|
| 712 |
+
1.0,
|
| 713 |
+
0.9999997615814209,
|
| 714 |
+
0.9999909400939941,
|
| 715 |
+
0.9995943903923035,
|
| 716 |
+
0.9999872446060181,
|
| 717 |
+
0.9999252557754517,
|
| 718 |
+
0.999969482421875,
|
| 719 |
+
0.9999967813491821,
|
| 720 |
+
0.9999842643737793,
|
| 721 |
+
0.9999707937240601,
|
| 722 |
+
0.9999200105667114,
|
| 723 |
+
0.9751906991004944,
|
| 724 |
+
0.10745485126972198,
|
| 725 |
+
0.9999172687530518,
|
| 726 |
+
0.9980173110961914,
|
| 727 |
+
0.964015007019043,
|
| 728 |
+
0.9997988343238831,
|
| 729 |
+
0.9999998807907104,
|
| 730 |
+
0.9999998807907104,
|
| 731 |
+
1.0,
|
| 732 |
+
1.0,
|
| 733 |
+
0.9999996423721313,
|
| 734 |
+
0.9999997615814209,
|
| 735 |
+
1.0,
|
| 736 |
+
1.0,
|
| 737 |
+
1.0,
|
| 738 |
+
0.9998918771743774,
|
| 739 |
+
0.9999985694885254,
|
| 740 |
+
0.9999735355377197,
|
| 741 |
+
0.9980869293212891,
|
| 742 |
+
0.8544897437095642,
|
| 743 |
+
0.9932467937469482,
|
| 744 |
+
0.999996542930603,
|
| 745 |
+
0.9999991655349731,
|
| 746 |
+
0.9999991655349731,
|
| 747 |
+
1.0,
|
| 748 |
+
1.0,
|
| 749 |
+
1.0,
|
| 750 |
+
0.9998795986175537,
|
| 751 |
+
0.9999253749847412,
|
| 752 |
+
0.9998786449432373,
|
| 753 |
+
1.0,
|
| 754 |
+
1.0,
|
| 755 |
+
1.0,
|
| 756 |
+
0.9999792575836182,
|
| 757 |
+
0.9999995231628418,
|
| 758 |
+
0.9999175071716309,
|
| 759 |
+
0.9999997615814209,
|
| 760 |
+
1.0,
|
| 761 |
+
0.9999998807907104,
|
| 762 |
+
0.9999803304672241,
|
| 763 |
+
0.9999405145645142,
|
| 764 |
+
0.9999480247497559,
|
| 765 |
+
0.9999998807907104,
|
| 766 |
+
1.0,
|
| 767 |
+
0.9999001026153564,
|
| 768 |
+
1.0,
|
| 769 |
+
1.0,
|
| 770 |
+
1.0,
|
| 771 |
+
1.0,
|
| 772 |
+
0.9999997615814209,
|
| 773 |
+
1.0,
|
| 774 |
+
0.9999985694885254,
|
| 775 |
+
0.9999977350234985,
|
| 776 |
+
0.9999935626983643,
|
| 777 |
+
0.9955407977104187,
|
| 778 |
+
0.29453033208847046,
|
| 779 |
+
0.9999270439147949,
|
| 780 |
+
0.9999998807907104,
|
| 781 |
+
1.0,
|
| 782 |
+
0.999995231628418,
|
| 783 |
+
0.265657901763916,
|
| 784 |
+
0.022009721025824547,
|
| 785 |
+
0.26402267813682556,
|
| 786 |
+
0.9994683861732483
|
| 787 |
+
],
|
| 788 |
+
"eval_math_textbook_runtime": 88.0387,
|
| 789 |
+
"eval_math_textbook_samples_per_second": 1.136,
|
| 790 |
+
"eval_math_textbook_steps_per_second": 0.045,
|
| 791 |
+
"epoch": 1.32,
|
| 792 |
+
"step": 75
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"loss": 0.0064,
|
| 796 |
+
"learning_rate": 0.0002,
|
| 797 |
+
"epoch": 1.33,
|
| 798 |
+
"step": 76
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"loss": 0.0032,
|
| 802 |
+
"learning_rate": 0.0002,
|
| 803 |
+
"epoch": 1.35,
|
| 804 |
+
"step": 77
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"loss": 0.0048,
|
| 808 |
+
"learning_rate": 0.0002,
|
| 809 |
+
"epoch": 1.37,
|
| 810 |
+
"step": 78
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"loss": 0.0003,
|
| 814 |
+
"learning_rate": 0.0002,
|
| 815 |
+
"epoch": 1.39,
|
| 816 |
+
"step": 79
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"loss": 0.0372,
|
| 820 |
+
"learning_rate": 0.0002,
|
| 821 |
+
"epoch": 1.4,
|
| 822 |
+
"step": 80
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"loss": 0.0939,
|
| 826 |
+
"learning_rate": 0.0002,
|
| 827 |
+
"epoch": 1.42,
|
| 828 |
+
"step": 81
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"loss": 0.0126,
|
| 832 |
+
"learning_rate": 0.0002,
|
| 833 |
+
"epoch": 1.44,
|
| 834 |
+
"step": 82
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"loss": 0.0108,
|
| 838 |
+
"learning_rate": 0.0002,
|
| 839 |
+
"epoch": 1.46,
|
| 840 |
+
"step": 83
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"loss": 0.0085,
|
| 844 |
+
"learning_rate": 0.0002,
|
| 845 |
+
"epoch": 1.47,
|
| 846 |
+
"step": 84
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"loss": 0.0208,
|
| 850 |
+
"learning_rate": 0.0002,
|
| 851 |
+
"epoch": 1.49,
|
| 852 |
+
"step": 85
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"loss": 0.0148,
|
| 856 |
+
"learning_rate": 0.0002,
|
| 857 |
+
"epoch": 1.51,
|
| 858 |
+
"step": 86
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"loss": 0.0621,
|
| 862 |
+
"learning_rate": 0.0002,
|
| 863 |
+
"epoch": 1.53,
|
| 864 |
+
"step": 87
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"loss": 0.0206,
|
| 868 |
+
"learning_rate": 0.0002,
|
| 869 |
+
"epoch": 1.54,
|
| 870 |
+
"step": 88
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"loss": 0.3222,
|
| 874 |
+
"learning_rate": 0.0002,
|
| 875 |
+
"epoch": 1.56,
|
| 876 |
+
"step": 89
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"loss": 0.0032,
|
| 880 |
+
"learning_rate": 0.0002,
|
| 881 |
+
"epoch": 1.58,
|
| 882 |
+
"step": 90
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"loss": 0.3471,
|
| 886 |
+
"learning_rate": 0.0002,
|
| 887 |
+
"epoch": 1.6,
|
| 888 |
+
"step": 91
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"loss": 0.0071,
|
| 892 |
+
"learning_rate": 0.0002,
|
| 893 |
+
"epoch": 1.61,
|
| 894 |
+
"step": 92
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"loss": 0.0237,
|
| 898 |
+
"learning_rate": 0.0002,
|
| 899 |
+
"epoch": 1.63,
|
| 900 |
+
"step": 93
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"loss": 0.001,
|
| 904 |
+
"learning_rate": 0.0002,
|
| 905 |
+
"epoch": 1.65,
|
| 906 |
+
"step": 94
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"loss": 0.0001,
|
| 910 |
+
"learning_rate": 0.0002,
|
| 911 |
+
"epoch": 1.67,
|
| 912 |
+
"step": 95
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"loss": 0.0929,
|
| 916 |
+
"learning_rate": 0.0002,
|
| 917 |
+
"epoch": 1.68,
|
| 918 |
+
"step": 96
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"loss": 0.0143,
|
| 922 |
+
"learning_rate": 0.0002,
|
| 923 |
+
"epoch": 1.7,
|
| 924 |
+
"step": 97
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"loss": 0.3065,
|
| 928 |
+
"learning_rate": 0.0002,
|
| 929 |
+
"epoch": 1.72,
|
| 930 |
+
"step": 98
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"loss": 0.0001,
|
| 934 |
+
"learning_rate": 0.0002,
|
| 935 |
+
"epoch": 1.74,
|
| 936 |
+
"step": 99
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"loss": 0.0032,
|
| 940 |
+
"learning_rate": 0.0002,
|
| 941 |
+
"epoch": 1.75,
|
| 942 |
+
"step": 100
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"eval_math_textbook_loss": 0.010601536370813847,
|
| 946 |
+
"eval_math_textbook_score": -0.002995953895151615,
|
| 947 |
+
"eval_math_textbook_brier_score": 0.002995953895151615,
|
| 948 |
+
"eval_math_textbook_average_probability": 0.988393247127533,
|
| 949 |
+
"eval_math_textbook_accuracy": 1.0,
|
| 950 |
+
"eval_math_textbook_probabilities": [
|
| 951 |
+
0.9999728202819824,
|
| 952 |
+
0.999929666519165,
|
| 953 |
+
0.999992847442627,
|
| 954 |
+
1.0,
|
| 955 |
+
1.0,
|
| 956 |
+
1.0,
|
| 957 |
+
0.972246527671814,
|
| 958 |
+
0.9996466636657715,
|
| 959 |
+
0.9953646659851074,
|
| 960 |
+
0.9999886751174927,
|
| 961 |
+
0.9970952272415161,
|
| 962 |
+
1.0,
|
| 963 |
+
1.0,
|
| 964 |
+
1.0,
|
| 965 |
+
1.0,
|
| 966 |
+
0.9495363831520081,
|
| 967 |
+
0.9999958276748657,
|
| 968 |
+
0.9999847412109375,
|
| 969 |
+
0.9338744878768921,
|
| 970 |
+
0.9645399451255798,
|
| 971 |
+
0.9999959468841553,
|
| 972 |
+
0.7106530070304871,
|
| 973 |
+
1.0,
|
| 974 |
+
1.0,
|
| 975 |
+
1.0,
|
| 976 |
+
1.0,
|
| 977 |
+
1.0,
|
| 978 |
+
0.9998949766159058,
|
| 979 |
+
0.9981856942176819,
|
| 980 |
+
0.9998927116394043,
|
| 981 |
+
0.9999417066574097,
|
| 982 |
+
0.9999963045120239,
|
| 983 |
+
0.9999362230300903,
|
| 984 |
+
0.9999959468841553,
|
| 985 |
+
0.9999982118606567,
|
| 986 |
+
0.9999841451644897,
|
| 987 |
+
0.9837162494659424,
|
| 988 |
+
0.9520687460899353,
|
| 989 |
+
0.9999997615814209,
|
| 990 |
+
0.999528169631958,
|
| 991 |
+
0.9320865273475647,
|
| 992 |
+
0.993480920791626,
|
| 993 |
+
0.9999996423721313,
|
| 994 |
+
0.9999998807907104,
|
| 995 |
+
0.9999997615814209,
|
| 996 |
+
0.9999998807907104,
|
| 997 |
+
0.9999971389770508,
|
| 998 |
+
0.9999990463256836,
|
| 999 |
+
1.0,
|
| 1000 |
+
1.0,
|
| 1001 |
+
1.0,
|
| 1002 |
+
0.9999992847442627,
|
| 1003 |
+
1.0,
|
| 1004 |
+
0.9999998807907104,
|
| 1005 |
+
0.9999986886978149,
|
| 1006 |
+
0.9849389791488647,
|
| 1007 |
+
0.9999982118606567,
|
| 1008 |
+
0.9999945163726807,
|
| 1009 |
+
0.999998927116394,
|
| 1010 |
+
0.9999977350234985,
|
| 1011 |
+
1.0,
|
| 1012 |
+
1.0,
|
| 1013 |
+
0.9999998807907104,
|
| 1014 |
+
1.0,
|
| 1015 |
+
1.0,
|
| 1016 |
+
1.0,
|
| 1017 |
+
1.0,
|
| 1018 |
+
1.0,
|
| 1019 |
+
1.0,
|
| 1020 |
+
0.9998576641082764,
|
| 1021 |
+
0.9999946355819702,
|
| 1022 |
+
0.9998277425765991,
|
| 1023 |
+
1.0,
|
| 1024 |
+
1.0,
|
| 1025 |
+
1.0,
|
| 1026 |
+
1.0,
|
| 1027 |
+
1.0,
|
| 1028 |
+
1.0,
|
| 1029 |
+
1.0,
|
| 1030 |
+
1.0,
|
| 1031 |
+
0.9998338222503662,
|
| 1032 |
+
0.9999983310699463,
|
| 1033 |
+
0.9999959468841553,
|
| 1034 |
+
0.9999998807907104,
|
| 1035 |
+
0.9995096921920776,
|
| 1036 |
+
0.9973734617233276,
|
| 1037 |
+
0.999901294708252,
|
| 1038 |
+
0.9999998807907104,
|
| 1039 |
+
0.9999998807907104,
|
| 1040 |
+
0.9999985694885254,
|
| 1041 |
+
0.993060827255249,
|
| 1042 |
+
0.9994891881942749,
|
| 1043 |
+
0.9998596906661987,
|
| 1044 |
+
0.9999996423721313,
|
| 1045 |
+
0.9999998807907104,
|
| 1046 |
+
0.9999985694885254,
|
| 1047 |
+
0.9552987217903137,
|
| 1048 |
+
0.5565717220306396,
|
| 1049 |
+
0.9723117351531982,
|
| 1050 |
+
0.9999910593032837
|
| 1051 |
+
],
|
| 1052 |
+
"eval_math_textbook_runtime": 87.9871,
|
| 1053 |
+
"eval_math_textbook_samples_per_second": 1.137,
|
| 1054 |
+
"eval_math_textbook_steps_per_second": 0.045,
|
| 1055 |
+
"epoch": 1.75,
|
| 1056 |
+
"step": 100
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"train_runtime": 7888.9918,
|
| 1060 |
+
"train_samples_per_second": 0.406,
|
| 1061 |
+
"train_steps_per_second": 0.013,
|
| 1062 |
+
"total_flos": 0.0,
|
| 1063 |
+
"train_loss": 0.19992217444516427,
|
| 1064 |
+
"epoch": 1.75,
|
| 1065 |
+
"step": 100
|
| 1066 |
+
}
|
| 1067 |
+
]
|