Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,64 @@
+---
+library_name: transformers
+license: other
+base_model: Qwen/Qwen2.5-Coder-7B-Instruct
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: MATH_training_response_Qwen2.5-32B-Instruct_common_correct_level
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# MATH_training_response_Qwen2.5-32B-Instruct_common_correct_level
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct) on the MATH_training_response_Qwen2.5-32B-Instruct_common_correct_level dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1231
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 4
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- total_train_batch_size: 16
+- total_eval_batch_size: 4
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- num_epochs: 2
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 0.0233        | 1.3699 | 200  | 0.1278          |
+
+
+### Framework versions
+
+- Transformers 4.46.1
+- Pytorch 2.5.1+cu124
+- Datasets 3.1.0
+- Tokenizers 0.20.3

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.0,
+    "eval_loss": 0.12309124320745468,
+    "eval_runtime": 3.5999,
+    "eval_samples_per_second": 6.667,
+    "eval_steps_per_second": 1.667,
+    "total_flos": 8378343751680.0,
+    "train_loss": 0.06280981206449948,
+    "train_runtime": 833.6474,
+    "train_samples_per_second": 5.587,
+    "train_steps_per_second": 0.35
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-Coder-7B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.1",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.0,
+    "eval_loss": 0.12309124320745468,
+    "eval_runtime": 3.5999,
+    "eval_samples_per_second": 6.667,
+    "eval_steps_per_second": 1.667
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.46.1"
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e9939398160d6ad810e731cc2b4292985cb3c74ab09f8451720ad81390a526b
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe3129555d733f8e81fe739766f91e2d7e0e679f887b957680c04a467ac9e18c
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d44c599db585cbb83c1a0bb90e09e4657297f93315ed6d050742bc632db2f68f
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcbeda28782a275fe7ddc5cf500462b99c0ed632ca46648a9dd1fe34d58d31f6
+size 1089994880

model.safetensors.index.json

@@ -0,0 +1,346 @@
+{
+  "metadata": {
+    "total_size": 15231233024
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00003-of-00004.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.0,
+    "total_flos": 8378343751680.0,
+    "train_loss": 0.06280981206449948,
+    "train_runtime": 833.6474,
+    "train_samples_per_second": 5.587,
+    "train_steps_per_second": 0.35
+}

trainer_log.jsonl

@@ -0,0 +1,294 @@
+{"current_steps": 1, "total_steps": 292, "loss": 0.1231, "lr": 9.999710619100732e-06, "epoch": 0.00684931506849315, "percentage": 0.34, "elapsed_time": "0:00:04", "remaining_time": "0:20:00"}
+{"current_steps": 2, "total_steps": 292, "loss": 0.1599, "lr": 9.998842509899456e-06, "epoch": 0.0136986301369863, "percentage": 0.68, "elapsed_time": "0:00:06", "remaining_time": "0:16:16"}
+{"current_steps": 3, "total_steps": 292, "loss": 0.1042, "lr": 9.997395772881853e-06, "epoch": 0.02054794520547945, "percentage": 1.03, "elapsed_time": "0:00:09", "remaining_time": "0:15:03"}
+{"current_steps": 4, "total_steps": 292, "loss": 0.114, "lr": 9.995370575511151e-06, "epoch": 0.0273972602739726, "percentage": 1.37, "elapsed_time": "0:00:11", "remaining_time": "0:14:20"}
+{"current_steps": 5, "total_steps": 292, "loss": 0.0993, "lr": 9.992767152208724e-06, "epoch": 0.03424657534246575, "percentage": 1.71, "elapsed_time": "0:00:14", "remaining_time": "0:14:12"}
+{"current_steps": 6, "total_steps": 292, "loss": 0.0887, "lr": 9.989585804326963e-06, "epoch": 0.0410958904109589, "percentage": 2.05, "elapsed_time": "0:00:17", "remaining_time": "0:13:55"}
+{"current_steps": 7, "total_steps": 292, "loss": 0.0992, "lr": 9.985826900114391e-06, "epoch": 0.04794520547945205, "percentage": 2.4, "elapsed_time": "0:00:19", "remaining_time": "0:13:33"}
+{"current_steps": 8, "total_steps": 292, "loss": 0.0936, "lr": 9.98149087467304e-06, "epoch": 0.0547945205479452, "percentage": 2.74, "elapsed_time": "0:00:22", "remaining_time": "0:13:19"}
+{"current_steps": 9, "total_steps": 292, "loss": 0.1073, "lr": 9.97657822990809e-06, "epoch": 0.06164383561643835, "percentage": 3.08, "elapsed_time": "0:00:25", "remaining_time": "0:13:11"}
+{"current_steps": 10, "total_steps": 292, "loss": 0.0993, "lr": 9.97108953446976e-06, "epoch": 0.0684931506849315, "percentage": 3.42, "elapsed_time": "0:00:27", "remaining_time": "0:13:04"}
+{"current_steps": 11, "total_steps": 292, "loss": 0.1084, "lr": 9.965025423687505e-06, "epoch": 0.07534246575342465, "percentage": 3.77, "elapsed_time": "0:00:30", "remaining_time": "0:12:55"}
+{"current_steps": 12, "total_steps": 292, "loss": 0.0843, "lr": 9.95838659949645e-06, "epoch": 0.0821917808219178, "percentage": 4.11, "elapsed_time": "0:00:32", "remaining_time": "0:12:49"}
+{"current_steps": 13, "total_steps": 292, "loss": 0.0894, "lr": 9.951173830356168e-06, "epoch": 0.08904109589041095, "percentage": 4.45, "elapsed_time": "0:00:35", "remaining_time": "0:12:47"}
+{"current_steps": 14, "total_steps": 292, "loss": 0.1161, "lr": 9.943387951161702e-06, "epoch": 0.0958904109589041, "percentage": 4.79, "elapsed_time": "0:00:38", "remaining_time": "0:12:39"}
+{"current_steps": 15, "total_steps": 292, "loss": 0.0872, "lr": 9.935029863146946e-06, "epoch": 0.10273972602739725, "percentage": 5.14, "elapsed_time": "0:00:40", "remaining_time": "0:12:33"}
+{"current_steps": 16, "total_steps": 292, "loss": 0.0697, "lr": 9.926100533780304e-06, "epoch": 0.1095890410958904, "percentage": 5.48, "elapsed_time": "0:00:43", "remaining_time": "0:12:34"}
+{"current_steps": 17, "total_steps": 292, "loss": 0.1001, "lr": 9.916600996652726e-06, "epoch": 0.11643835616438356, "percentage": 5.82, "elapsed_time": "0:00:46", "remaining_time": "0:12:27"}
+{"current_steps": 18, "total_steps": 292, "loss": 0.1142, "lr": 9.906532351358047e-06, "epoch": 0.1232876712328767, "percentage": 6.16, "elapsed_time": "0:00:48", "remaining_time": "0:12:22"}
+{"current_steps": 19, "total_steps": 292, "loss": 0.0951, "lr": 9.895895763365722e-06, "epoch": 0.13013698630136986, "percentage": 6.51, "elapsed_time": "0:00:51", "remaining_time": "0:12:18"}
+{"current_steps": 20, "total_steps": 292, "loss": 0.1102, "lr": 9.88469246388591e-06, "epoch": 0.136986301369863, "percentage": 6.85, "elapsed_time": "0:00:54", "remaining_time": "0:12:15"}
+{"current_steps": 21, "total_steps": 292, "loss": 0.1102, "lr": 9.872923749726959e-06, "epoch": 0.14383561643835616, "percentage": 7.19, "elapsed_time": "0:00:56", "remaining_time": "0:12:10"}
+{"current_steps": 22, "total_steps": 292, "loss": 0.0889, "lr": 9.860590983145307e-06, "epoch": 0.1506849315068493, "percentage": 7.53, "elapsed_time": "0:00:59", "remaining_time": "0:12:08"}
+{"current_steps": 23, "total_steps": 292, "loss": 0.1128, "lr": 9.847695591687788e-06, "epoch": 0.15753424657534246, "percentage": 7.88, "elapsed_time": "0:01:01", "remaining_time": "0:12:04"}
+{"current_steps": 24, "total_steps": 292, "loss": 0.099, "lr": 9.834239068026388e-06, "epoch": 0.1643835616438356, "percentage": 8.22, "elapsed_time": "0:01:04", "remaining_time": "0:12:02"}
+{"current_steps": 25, "total_steps": 292, "loss": 0.0931, "lr": 9.82022296978548e-06, "epoch": 0.17123287671232876, "percentage": 8.56, "elapsed_time": "0:01:07", "remaining_time": "0:11:57"}
+{"current_steps": 26, "total_steps": 292, "loss": 0.0875, "lr": 9.805648919361505e-06, "epoch": 0.1780821917808219, "percentage": 8.9, "elapsed_time": "0:01:09", "remaining_time": "0:11:53"}
+{"current_steps": 27, "total_steps": 292, "loss": 0.1036, "lr": 9.790518603735191e-06, "epoch": 0.18493150684931506, "percentage": 9.25, "elapsed_time": "0:01:12", "remaining_time": "0:11:48"}
+{"current_steps": 28, "total_steps": 292, "loss": 0.0994, "lr": 9.774833774276278e-06, "epoch": 0.1917808219178082, "percentage": 9.59, "elapsed_time": "0:01:14", "remaining_time": "0:11:44"}
+{"current_steps": 29, "total_steps": 292, "loss": 0.1192, "lr": 9.758596246540782e-06, "epoch": 0.19863013698630136, "percentage": 9.93, "elapsed_time": "0:01:17", "remaining_time": "0:11:41"}
+{"current_steps": 30, "total_steps": 292, "loss": 0.0928, "lr": 9.741807900060858e-06, "epoch": 0.2054794520547945, "percentage": 10.27, "elapsed_time": "0:01:19", "remaining_time": "0:11:36"}
+{"current_steps": 31, "total_steps": 292, "loss": 0.1115, "lr": 9.724470678127226e-06, "epoch": 0.21232876712328766, "percentage": 10.62, "elapsed_time": "0:01:22", "remaining_time": "0:11:34"}
+{"current_steps": 32, "total_steps": 292, "loss": 0.109, "lr": 9.706586587564236e-06, "epoch": 0.2191780821917808, "percentage": 10.96, "elapsed_time": "0:01:25", "remaining_time": "0:11:30"}
+{"current_steps": 33, "total_steps": 292, "loss": 0.1025, "lr": 9.68815769849757e-06, "epoch": 0.22602739726027396, "percentage": 11.3, "elapsed_time": "0:01:27", "remaining_time": "0:11:27"}
+{"current_steps": 34, "total_steps": 292, "loss": 0.1096, "lr": 9.669186144114627e-06, "epoch": 0.2328767123287671, "percentage": 11.64, "elapsed_time": "0:01:30", "remaining_time": "0:11:23"}
+{"current_steps": 35, "total_steps": 292, "loss": 0.1092, "lr": 9.649674120417591e-06, "epoch": 0.23972602739726026, "percentage": 11.99, "elapsed_time": "0:01:32", "remaining_time": "0:11:20"}
+{"current_steps": 36, "total_steps": 292, "loss": 0.1106, "lr": 9.62962388596925e-06, "epoch": 0.2465753424657534, "percentage": 12.33, "elapsed_time": "0:01:35", "remaining_time": "0:11:16"}
+{"current_steps": 37, "total_steps": 292, "loss": 0.1164, "lr": 9.609037761631552e-06, "epoch": 0.2534246575342466, "percentage": 12.67, "elapsed_time": "0:01:37", "remaining_time": "0:11:13"}
+{"current_steps": 38, "total_steps": 292, "loss": 0.1041, "lr": 9.587918130296969e-06, "epoch": 0.2602739726027397, "percentage": 13.01, "elapsed_time": "0:01:40", "remaining_time": "0:11:09"}
+{"current_steps": 39, "total_steps": 292, "loss": 0.0961, "lr": 9.566267436612662e-06, "epoch": 0.2671232876712329, "percentage": 13.36, "elapsed_time": "0:01:42", "remaining_time": "0:11:07"}
+{"current_steps": 40, "total_steps": 292, "loss": 0.1033, "lr": 9.544088186697515e-06, "epoch": 0.273972602739726, "percentage": 13.7, "elapsed_time": "0:01:45", "remaining_time": "0:11:05"}
+{"current_steps": 41, "total_steps": 292, "loss": 0.0692, "lr": 9.521382947852042e-06, "epoch": 0.2808219178082192, "percentage": 14.04, "elapsed_time": "0:01:48", "remaining_time": "0:11:01"}
+{"current_steps": 42, "total_steps": 292, "loss": 0.1024, "lr": 9.498154348261217e-06, "epoch": 0.2876712328767123, "percentage": 14.38, "elapsed_time": "0:01:50", "remaining_time": "0:10:59"}
+{"current_steps": 43, "total_steps": 292, "loss": 0.1045, "lr": 9.474405076690257e-06, "epoch": 0.2945205479452055, "percentage": 14.73, "elapsed_time": "0:01:53", "remaining_time": "0:10:56"}
+{"current_steps": 44, "total_steps": 292, "loss": 0.088, "lr": 9.450137882173385e-06, "epoch": 0.3013698630136986, "percentage": 15.07, "elapsed_time": "0:01:56", "remaining_time": "0:10:54"}
+{"current_steps": 45, "total_steps": 292, "loss": 0.089, "lr": 9.425355573695628e-06, "epoch": 0.3082191780821918, "percentage": 15.41, "elapsed_time": "0:01:58", "remaining_time": "0:10:52"}
+{"current_steps": 46, "total_steps": 292, "loss": 0.1076, "lr": 9.40006101986768e-06, "epoch": 0.3150684931506849, "percentage": 15.75, "elapsed_time": "0:02:01", "remaining_time": "0:10:49"}
+{"current_steps": 47, "total_steps": 292, "loss": 0.0997, "lr": 9.374257148593824e-06, "epoch": 0.3219178082191781, "percentage": 16.1, "elapsed_time": "0:02:04", "remaining_time": "0:10:47"}
+{"current_steps": 48, "total_steps": 292, "loss": 0.1089, "lr": 9.347946946733055e-06, "epoch": 0.3287671232876712, "percentage": 16.44, "elapsed_time": "0:02:06", "remaining_time": "0:10:44"}
+{"current_steps": 49, "total_steps": 292, "loss": 0.1028, "lr": 9.321133459753322e-06, "epoch": 0.3356164383561644, "percentage": 16.78, "elapsed_time": "0:02:09", "remaining_time": "0:10:41"}
+{"current_steps": 50, "total_steps": 292, "loss": 0.1166, "lr": 9.293819791379016e-06, "epoch": 0.3424657534246575, "percentage": 17.12, "elapsed_time": "0:02:11", "remaining_time": "0:10:38"}
+{"current_steps": 51, "total_steps": 292, "loss": 0.1008, "lr": 9.266009103231702e-06, "epoch": 0.3493150684931507, "percentage": 17.47, "elapsed_time": "0:02:14", "remaining_time": "0:10:35"}
+{"current_steps": 52, "total_steps": 292, "loss": 0.1126, "lr": 9.237704614464157e-06, "epoch": 0.3561643835616438, "percentage": 17.81, "elapsed_time": "0:02:16", "remaining_time": "0:10:31"}
+{"current_steps": 53, "total_steps": 292, "loss": 0.1081, "lr": 9.208909601387748e-06, "epoch": 0.363013698630137, "percentage": 18.15, "elapsed_time": "0:02:19", "remaining_time": "0:10:28"}
+{"current_steps": 54, "total_steps": 292, "loss": 0.0901, "lr": 9.179627397093184e-06, "epoch": 0.3698630136986301, "percentage": 18.49, "elapsed_time": "0:02:21", "remaining_time": "0:10:24"}
+{"current_steps": 55, "total_steps": 292, "loss": 0.1017, "lr": 9.149861391064714e-06, "epoch": 0.3767123287671233, "percentage": 18.84, "elapsed_time": "0:02:24", "remaining_time": "0:10:22"}
+{"current_steps": 56, "total_steps": 292, "loss": 0.1018, "lr": 9.119615028787771e-06, "epoch": 0.3835616438356164, "percentage": 19.18, "elapsed_time": "0:02:27", "remaining_time": "0:10:20"}
+{"current_steps": 57, "total_steps": 292, "loss": 0.1049, "lr": 9.088891811350164e-06, "epoch": 0.3904109589041096, "percentage": 19.52, "elapsed_time": "0:02:29", "remaining_time": "0:10:17"}
+{"current_steps": 58, "total_steps": 292, "loss": 0.1042, "lr": 9.057695295036806e-06, "epoch": 0.3972602739726027, "percentage": 19.86, "elapsed_time": "0:02:32", "remaining_time": "0:10:14"}
+{"current_steps": 59, "total_steps": 292, "loss": 0.0888, "lr": 9.026029090918076e-06, "epoch": 0.4041095890410959, "percentage": 20.21, "elapsed_time": "0:02:35", "remaining_time": "0:10:12"}
+{"current_steps": 60, "total_steps": 292, "loss": 0.1002, "lr": 8.993896864431825e-06, "epoch": 0.410958904109589, "percentage": 20.55, "elapsed_time": "0:02:37", "remaining_time": "0:10:09"}
+{"current_steps": 61, "total_steps": 292, "loss": 0.0881, "lr": 8.96130233495909e-06, "epoch": 0.4178082191780822, "percentage": 20.89, "elapsed_time": "0:02:40", "remaining_time": "0:10:06"}
+{"current_steps": 62, "total_steps": 292, "loss": 0.0944, "lr": 8.928249275393572e-06, "epoch": 0.4246575342465753, "percentage": 21.23, "elapsed_time": "0:02:43", "remaining_time": "0:10:04"}
+{"current_steps": 63, "total_steps": 292, "loss": 0.0874, "lr": 8.894741511704911e-06, "epoch": 0.4315068493150685, "percentage": 21.58, "elapsed_time": "0:02:45", "remaining_time": "0:10:02"}
+{"current_steps": 64, "total_steps": 292, "loss": 0.1053, "lr": 8.860782922495821e-06, "epoch": 0.4383561643835616, "percentage": 21.92, "elapsed_time": "0:02:48", "remaining_time": "0:10:00"}
+{"current_steps": 65, "total_steps": 292, "loss": 0.1098, "lr": 8.826377438553138e-06, "epoch": 0.4452054794520548, "percentage": 22.26, "elapsed_time": "0:02:51", "remaining_time": "0:09:57"}
+{"current_steps": 66, "total_steps": 292, "loss": 0.0983, "lr": 8.791529042392813e-06, "epoch": 0.4520547945205479, "percentage": 22.6, "elapsed_time": "0:02:53", "remaining_time": "0:09:54"}
+{"current_steps": 67, "total_steps": 292, "loss": 0.0922, "lr": 8.756241767798934e-06, "epoch": 0.4589041095890411, "percentage": 22.95, "elapsed_time": "0:02:56", "remaining_time": "0:09:51"}
+{"current_steps": 68, "total_steps": 292, "loss": 0.1009, "lr": 8.720519699356804e-06, "epoch": 0.4657534246575342, "percentage": 23.29, "elapsed_time": "0:02:58", "remaining_time": "0:09:48"}
+{"current_steps": 69, "total_steps": 292, "loss": 0.1136, "lr": 8.684366971980139e-06, "epoch": 0.4726027397260274, "percentage": 23.63, "elapsed_time": "0:03:01", "remaining_time": "0:09:45"}
+{"current_steps": 70, "total_steps": 292, "loss": 0.106, "lr": 8.647787770432439e-06, "epoch": 0.4794520547945205, "percentage": 23.97, "elapsed_time": "0:03:04", "remaining_time": "0:09:43"}
+{"current_steps": 71, "total_steps": 292, "loss": 0.0781, "lr": 8.610786328842602e-06, "epoch": 0.4863013698630137, "percentage": 24.32, "elapsed_time": "0:03:06", "remaining_time": "0:09:41"}
+{"current_steps": 72, "total_steps": 292, "loss": 0.0953, "lr": 8.573366930214807e-06, "epoch": 0.4931506849315068, "percentage": 24.66, "elapsed_time": "0:03:09", "remaining_time": "0:09:38"}
+{"current_steps": 73, "total_steps": 292, "loss": 0.0977, "lr": 8.535533905932739e-06, "epoch": 0.5, "percentage": 25.0, "elapsed_time": "0:03:11", "remaining_time": "0:09:35"}
+{"current_steps": 74, "total_steps": 292, "loss": 0.0986, "lr": 8.497291635258235e-06, "epoch": 0.5068493150684932, "percentage": 25.34, "elapsed_time": "0:03:14", "remaining_time": "0:09:33"}
+{"current_steps": 75, "total_steps": 292, "loss": 0.1066, "lr": 8.458644544824371e-06, "epoch": 0.5136986301369864, "percentage": 25.68, "elapsed_time": "0:03:16", "remaining_time": "0:09:29"}
+{"current_steps": 76, "total_steps": 292, "loss": 0.0778, "lr": 8.419597108123054e-06, "epoch": 0.5205479452054794, "percentage": 26.03, "elapsed_time": "0:03:19", "remaining_time": "0:09:27"}
+{"current_steps": 77, "total_steps": 292, "loss": 0.0885, "lr": 8.380153844987225e-06, "epoch": 0.5273972602739726, "percentage": 26.37, "elapsed_time": "0:03:22", "remaining_time": "0:09:24"}
+{"current_steps": 78, "total_steps": 292, "loss": 0.0784, "lr": 8.340319321067668e-06, "epoch": 0.5342465753424658, "percentage": 26.71, "elapsed_time": "0:03:24", "remaining_time": "0:09:22"}
+{"current_steps": 79, "total_steps": 292, "loss": 0.1051, "lr": 8.300098147304523e-06, "epoch": 0.541095890410959, "percentage": 27.05, "elapsed_time": "0:03:27", "remaining_time": "0:09:19"}
+{"current_steps": 80, "total_steps": 292, "loss": 0.1233, "lr": 8.259494979393563e-06, "epoch": 0.547945205479452, "percentage": 27.4, "elapsed_time": "0:03:30", "remaining_time": "0:09:16"}
+{"current_steps": 81, "total_steps": 292, "loss": 0.1009, "lr": 8.218514517247287e-06, "epoch": 0.5547945205479452, "percentage": 27.74, "elapsed_time": "0:03:32", "remaining_time": "0:09:14"}
+{"current_steps": 82, "total_steps": 292, "loss": 0.0909, "lr": 8.177161504450887e-06, "epoch": 0.5616438356164384, "percentage": 28.08, "elapsed_time": "0:03:35", "remaining_time": "0:09:11"}
+{"current_steps": 83, "total_steps": 292, "loss": 0.0965, "lr": 8.135440727713179e-06, "epoch": 0.5684931506849316, "percentage": 28.42, "elapsed_time": "0:03:37", "remaining_time": "0:09:07"}
+{"current_steps": 84, "total_steps": 292, "loss": 0.1382, "lr": 8.093357016312518e-06, "epoch": 0.5753424657534246, "percentage": 28.77, "elapsed_time": "0:03:40", "remaining_time": "0:09:04"}
+{"current_steps": 85, "total_steps": 292, "loss": 0.111, "lr": 8.050915241537802e-06, "epoch": 0.5821917808219178, "percentage": 29.11, "elapsed_time": "0:03:42", "remaining_time": "0:09:01"}
+{"current_steps": 86, "total_steps": 292, "loss": 0.0903, "lr": 8.008120316124612e-06, "epoch": 0.589041095890411, "percentage": 29.45, "elapsed_time": "0:03:45", "remaining_time": "0:08:59"}
+{"current_steps": 87, "total_steps": 292, "loss": 0.0889, "lr": 7.964977193686551e-06, "epoch": 0.5958904109589042, "percentage": 29.79, "elapsed_time": "0:03:47", "remaining_time": "0:08:56"}
+{"current_steps": 88, "total_steps": 292, "loss": 0.1025, "lr": 7.921490868141843e-06, "epoch": 0.6027397260273972, "percentage": 30.14, "elapsed_time": "0:03:50", "remaining_time": "0:08:53"}
+{"current_steps": 89, "total_steps": 292, "loss": 0.0872, "lr": 7.877666373135287e-06, "epoch": 0.6095890410958904, "percentage": 30.48, "elapsed_time": "0:03:52", "remaining_time": "0:08:50"}
+{"current_steps": 90, "total_steps": 292, "loss": 0.0976, "lr": 7.83350878145559e-06, "epoch": 0.6164383561643836, "percentage": 30.82, "elapsed_time": "0:03:55", "remaining_time": "0:08:48"}
+{"current_steps": 91, "total_steps": 292, "loss": 0.0829, "lr": 7.789023204448189e-06, "epoch": 0.6232876712328768, "percentage": 31.16, "elapsed_time": "0:03:58", "remaining_time": "0:08:46"}
+{"current_steps": 92, "total_steps": 292, "loss": 0.0862, "lr": 7.744214791423597e-06, "epoch": 0.6301369863013698, "percentage": 31.51, "elapsed_time": "0:04:01", "remaining_time": "0:08:44"}
+{"current_steps": 93, "total_steps": 292, "loss": 0.1227, "lr": 7.699088729061355e-06, "epoch": 0.636986301369863, "percentage": 31.85, "elapsed_time": "0:04:03", "remaining_time": "0:08:41"}
+{"current_steps": 94, "total_steps": 292, "loss": 0.0939, "lr": 7.653650240809667e-06, "epoch": 0.6438356164383562, "percentage": 32.19, "elapsed_time": "0:04:06", "remaining_time": "0:08:38"}
+{"current_steps": 95, "total_steps": 292, "loss": 0.0957, "lr": 7.60790458628077e-06, "epoch": 0.6506849315068494, "percentage": 32.53, "elapsed_time": "0:04:08", "remaining_time": "0:08:36"}
+{"current_steps": 96, "total_steps": 292, "loss": 0.0975, "lr": 7.56185706064212e-06, "epoch": 0.6575342465753424, "percentage": 32.88, "elapsed_time": "0:04:11", "remaining_time": "0:08:33"}
+{"current_steps": 97, "total_steps": 292, "loss": 0.0857, "lr": 7.5155129940034675e-06, "epoch": 0.6643835616438356, "percentage": 33.22, "elapsed_time": "0:04:14", "remaining_time": "0:08:31"}
+{"current_steps": 98, "total_steps": 292, "loss": 0.0995, "lr": 7.468877750799887e-06, "epoch": 0.6712328767123288, "percentage": 33.56, "elapsed_time": "0:04:17", "remaining_time": "0:08:29"}
+{"current_steps": 99, "total_steps": 292, "loss": 0.0779, "lr": 7.421956729170823e-06, "epoch": 0.678082191780822, "percentage": 33.9, "elapsed_time": "0:04:19", "remaining_time": "0:08:26"}
+{"current_steps": 100, "total_steps": 292, "loss": 0.0908, "lr": 7.374755360335253e-06, "epoch": 0.684931506849315, "percentage": 34.25, "elapsed_time": "0:04:22", "remaining_time": "0:08:23"}
+{"current_steps": 101, "total_steps": 292, "loss": 0.094, "lr": 7.327279107962995e-06, "epoch": 0.6917808219178082, "percentage": 34.59, "elapsed_time": "0:04:24", "remaining_time": "0:08:20"}
+{"current_steps": 102, "total_steps": 292, "loss": 0.1122, "lr": 7.279533467542295e-06, "epoch": 0.6986301369863014, "percentage": 34.93, "elapsed_time": "0:04:27", "remaining_time": "0:08:17"}
+{"current_steps": 103, "total_steps": 292, "loss": 0.094, "lr": 7.2315239657436955e-06, "epoch": 0.7054794520547946, "percentage": 35.27, "elapsed_time": "0:04:29", "remaining_time": "0:08:15"}
+{"current_steps": 104, "total_steps": 292, "loss": 0.1022, "lr": 7.183256159780321e-06, "epoch": 0.7123287671232876, "percentage": 35.62, "elapsed_time": "0:04:32", "remaining_time": "0:08:12"}
+{"current_steps": 105, "total_steps": 292, "loss": 0.0998, "lr": 7.134735636764606e-06, "epoch": 0.7191780821917808, "percentage": 35.96, "elapsed_time": "0:04:35", "remaining_time": "0:08:11"}
+{"current_steps": 106, "total_steps": 292, "loss": 0.1122, "lr": 7.085968013061585e-06, "epoch": 0.726027397260274, "percentage": 36.3, "elapsed_time": "0:04:38", "remaining_time": "0:08:08"}
+{"current_steps": 107, "total_steps": 292, "loss": 0.074, "lr": 7.036958933638779e-06, "epoch": 0.7328767123287672, "percentage": 36.64, "elapsed_time": "0:04:41", "remaining_time": "0:08:05"}
+{"current_steps": 108, "total_steps": 292, "loss": 0.1103, "lr": 6.987714071412781e-06, "epoch": 0.7397260273972602, "percentage": 36.99, "elapsed_time": "0:04:43", "remaining_time": "0:08:02"}
+{"current_steps": 109, "total_steps": 292, "loss": 0.1014, "lr": 6.938239126592592e-06, "epoch": 0.7465753424657534, "percentage": 37.33, "elapsed_time": "0:04:46", "remaining_time": "0:08:00"}
+{"current_steps": 110, "total_steps": 292, "loss": 0.0908, "lr": 6.888539826019824e-06, "epoch": 0.7534246575342466, "percentage": 37.67, "elapsed_time": "0:04:48", "remaining_time": "0:07:57"}
+{"current_steps": 111, "total_steps": 292, "loss": 0.0973, "lr": 6.8386219225057945e-06, "epoch": 0.7602739726027398, "percentage": 38.01, "elapsed_time": "0:04:51", "remaining_time": "0:07:54"}
+{"current_steps": 112, "total_steps": 292, "loss": 0.0983, "lr": 6.788491194165629e-06, "epoch": 0.7671232876712328, "percentage": 38.36, "elapsed_time": "0:04:53", "remaining_time": "0:07:52"}
+{"current_steps": 113, "total_steps": 292, "loss": 0.101, "lr": 6.738153443749421e-06, "epoch": 0.773972602739726, "percentage": 38.7, "elapsed_time": "0:04:56", "remaining_time": "0:07:49"}
+{"current_steps": 114, "total_steps": 292, "loss": 0.0949, "lr": 6.687614497970567e-06, "epoch": 0.7808219178082192, "percentage": 39.04, "elapsed_time": "0:04:59", "remaining_time": "0:07:47"}
+{"current_steps": 115, "total_steps": 292, "loss": 0.1011, "lr": 6.636880206831298e-06, "epoch": 0.7876712328767124, "percentage": 39.38, "elapsed_time": "0:05:01", "remaining_time": "0:07:44"}
+{"current_steps": 116, "total_steps": 292, "loss": 0.0872, "lr": 6.585956442945531e-06, "epoch": 0.7945205479452054, "percentage": 39.73, "elapsed_time": "0:05:04", "remaining_time": "0:07:41"}
+{"current_steps": 117, "total_steps": 292, "loss": 0.0859, "lr": 6.534849100859101e-06, "epoch": 0.8013698630136986, "percentage": 40.07, "elapsed_time": "0:05:07", "remaining_time": "0:07:40"}
+{"current_steps": 118, "total_steps": 292, "loss": 0.1142, "lr": 6.483564096367452e-06, "epoch": 0.8082191780821918, "percentage": 40.41, "elapsed_time": "0:05:10", "remaining_time": "0:07:37"}
+{"current_steps": 119, "total_steps": 292, "loss": 0.1047, "lr": 6.432107365830872e-06, "epoch": 0.815068493150685, "percentage": 40.75, "elapsed_time": "0:05:12", "remaining_time": "0:07:34"}
+{"current_steps": 120, "total_steps": 292, "loss": 0.1174, "lr": 6.380484865487346e-06, "epoch": 0.821917808219178, "percentage": 41.1, "elapsed_time": "0:05:15", "remaining_time": "0:07:32"}
+{"current_steps": 121, "total_steps": 292, "loss": 0.1014, "lr": 6.328702570763098e-06, "epoch": 0.8287671232876712, "percentage": 41.44, "elapsed_time": "0:05:17", "remaining_time": "0:07:29"}
+{"current_steps": 122, "total_steps": 292, "loss": 0.0888, "lr": 6.276766475580935e-06, "epoch": 0.8356164383561644, "percentage": 41.78, "elapsed_time": "0:05:20", "remaining_time": "0:07:26"}
+{"current_steps": 123, "total_steps": 292, "loss": 0.1194, "lr": 6.224682591666431e-06, "epoch": 0.8424657534246576, "percentage": 42.12, "elapsed_time": "0:05:22", "remaining_time": "0:07:23"}
+{"current_steps": 124, "total_steps": 292, "loss": 0.0766, "lr": 6.1724569478520495e-06, "epoch": 0.8493150684931506, "percentage": 42.47, "elapsed_time": "0:05:26", "remaining_time": "0:07:21"}
+{"current_steps": 125, "total_steps": 292, "loss": 0.0999, "lr": 6.120095589379299e-06, "epoch": 0.8561643835616438, "percentage": 42.81, "elapsed_time": "0:05:28", "remaining_time": "0:07:18"}
+{"current_steps": 126, "total_steps": 292, "loss": 0.1026, "lr": 6.067604577198981e-06, "epoch": 0.863013698630137, "percentage": 43.15, "elapsed_time": "0:05:31", "remaining_time": "0:07:17"}
+{"current_steps": 127, "total_steps": 292, "loss": 0.1177, "lr": 6.014989987269617e-06, "epoch": 0.8698630136986302, "percentage": 43.49, "elapsed_time": "0:05:34", "remaining_time": "0:07:14"}
+{"current_steps": 128, "total_steps": 292, "loss": 0.0841, "lr": 5.96225790985415e-06, "epoch": 0.8767123287671232, "percentage": 43.84, "elapsed_time": "0:05:36", "remaining_time": "0:07:11"}
+{"current_steps": 129, "total_steps": 292, "loss": 0.0668, "lr": 5.909414448814971e-06, "epoch": 0.8835616438356164, "percentage": 44.18, "elapsed_time": "0:05:39", "remaining_time": "0:07:08"}
+{"current_steps": 130, "total_steps": 292, "loss": 0.1175, "lr": 5.856465720907388e-06, "epoch": 0.8904109589041096, "percentage": 44.52, "elapsed_time": "0:05:41", "remaining_time": "0:07:05"}
+{"current_steps": 131, "total_steps": 292, "loss": 0.1083, "lr": 5.803417855071603e-06, "epoch": 0.8972602739726028, "percentage": 44.86, "elapsed_time": "0:05:44", "remaining_time": "0:07:03"}
+{"current_steps": 132, "total_steps": 292, "loss": 0.1048, "lr": 5.7502769917232635e-06, "epoch": 0.9041095890410958, "percentage": 45.21, "elapsed_time": "0:05:46", "remaining_time": "0:07:00"}
+{"current_steps": 133, "total_steps": 292, "loss": 0.0984, "lr": 5.6970492820426994e-06, "epoch": 0.910958904109589, "percentage": 45.55, "elapsed_time": "0:05:49", "remaining_time": "0:06:57"}
+{"current_steps": 134, "total_steps": 292, "loss": 0.0909, "lr": 5.643740887262905e-06, "epoch": 0.9178082191780822, "percentage": 45.89, "elapsed_time": "0:05:52", "remaining_time": "0:06:55"}
+{"current_steps": 135, "total_steps": 292, "loss": 0.1065, "lr": 5.59035797795637e-06, "epoch": 0.9246575342465754, "percentage": 46.23, "elapsed_time": "0:05:54", "remaining_time": "0:06:52"}
+{"current_steps": 136, "total_steps": 292, "loss": 0.0923, "lr": 5.536906733320816e-06, "epoch": 0.9315068493150684, "percentage": 46.58, "elapsed_time": "0:05:57", "remaining_time": "0:06:49"}
+{"current_steps": 137, "total_steps": 292, "loss": 0.0903, "lr": 5.483393340463938e-06, "epoch": 0.9383561643835616, "percentage": 46.92, "elapsed_time": "0:05:59", "remaining_time": "0:06:47"}
+{"current_steps": 138, "total_steps": 292, "loss": 0.1001, "lr": 5.429823993687234e-06, "epoch": 0.9452054794520548, "percentage": 47.26, "elapsed_time": "0:06:02", "remaining_time": "0:06:44"}
+{"current_steps": 139, "total_steps": 292, "loss": 0.0828, "lr": 5.376204893769e-06, "epoch": 0.952054794520548, "percentage": 47.6, "elapsed_time": "0:06:04", "remaining_time": "0:06:41"}
+{"current_steps": 140, "total_steps": 292, "loss": 0.0895, "lr": 5.322542247246583e-06, "epoch": 0.958904109589041, "percentage": 47.95, "elapsed_time": "0:06:07", "remaining_time": "0:06:38"}
+{"current_steps": 141, "total_steps": 292, "loss": 0.0877, "lr": 5.26884226569794e-06, "epoch": 0.9657534246575342, "percentage": 48.29, "elapsed_time": "0:06:10", "remaining_time": "0:06:36"}
+{"current_steps": 142, "total_steps": 292, "loss": 0.0935, "lr": 5.215111165022653e-06, "epoch": 0.9726027397260274, "percentage": 48.63, "elapsed_time": "0:06:12", "remaining_time": "0:06:33"}
+{"current_steps": 143, "total_steps": 292, "loss": 0.105, "lr": 5.161355164722416e-06, "epoch": 0.9794520547945206, "percentage": 48.97, "elapsed_time": "0:06:15", "remaining_time": "0:06:31"}
+{"current_steps": 144, "total_steps": 292, "loss": 0.0811, "lr": 5.107580487181112e-06, "epoch": 0.9863013698630136, "percentage": 49.32, "elapsed_time": "0:06:18", "remaining_time": "0:06:28"}
+{"current_steps": 145, "total_steps": 292, "loss": 0.0776, "lr": 5.0537933569445585e-06, "epoch": 0.9931506849315068, "percentage": 49.66, "elapsed_time": "0:06:21", "remaining_time": "0:06:26"}
+{"current_steps": 146, "total_steps": 292, "loss": 0.0813, "lr": 5e-06, "epoch": 1.0, "percentage": 50.0, "elapsed_time": "0:06:23", "remaining_time": "0:06:23"}
+{"current_steps": 147, "total_steps": 292, "loss": 0.035, "lr": 4.946206643055443e-06, "epoch": 1.0068493150684932, "percentage": 50.34, "elapsed_time": "0:06:26", "remaining_time": "0:06:21"}
+{"current_steps": 148, "total_steps": 292, "loss": 0.0242, "lr": 4.89241951281889e-06, "epoch": 1.0136986301369864, "percentage": 50.68, "elapsed_time": "0:06:29", "remaining_time": "0:06:18"}
+{"current_steps": 149, "total_steps": 292, "loss": 0.0393, "lr": 4.838644835277585e-06, "epoch": 1.0205479452054795, "percentage": 51.03, "elapsed_time": "0:06:31", "remaining_time": "0:06:15"}
+{"current_steps": 150, "total_steps": 292, "loss": 0.0299, "lr": 4.784888834977347e-06, "epoch": 1.0273972602739727, "percentage": 51.37, "elapsed_time": "0:06:34", "remaining_time": "0:06:13"}
+{"current_steps": 151, "total_steps": 292, "loss": 0.0408, "lr": 4.731157734302063e-06, "epoch": 1.0342465753424657, "percentage": 51.71, "elapsed_time": "0:06:36", "remaining_time": "0:06:10"}
+{"current_steps": 152, "total_steps": 292, "loss": 0.0294, "lr": 4.6774577527534195e-06, "epoch": 1.0410958904109588, "percentage": 52.05, "elapsed_time": "0:06:39", "remaining_time": "0:06:08"}
+{"current_steps": 153, "total_steps": 292, "loss": 0.0402, "lr": 4.623795106231001e-06, "epoch": 1.047945205479452, "percentage": 52.4, "elapsed_time": "0:06:42", "remaining_time": "0:06:05"}
+{"current_steps": 154, "total_steps": 292, "loss": 0.0306, "lr": 4.570176006312769e-06, "epoch": 1.0547945205479452, "percentage": 52.74, "elapsed_time": "0:06:44", "remaining_time": "0:06:02"}
+{"current_steps": 155, "total_steps": 292, "loss": 0.0325, "lr": 4.516606659536063e-06, "epoch": 1.0616438356164384, "percentage": 53.08, "elapsed_time": "0:06:47", "remaining_time": "0:06:00"}
+{"current_steps": 156, "total_steps": 292, "loss": 0.041, "lr": 4.463093266679185e-06, "epoch": 1.0684931506849316, "percentage": 53.42, "elapsed_time": "0:06:50", "remaining_time": "0:05:57"}
+{"current_steps": 157, "total_steps": 292, "loss": 0.0278, "lr": 4.40964202204363e-06, "epoch": 1.0753424657534247, "percentage": 53.77, "elapsed_time": "0:06:53", "remaining_time": "0:05:55"}
+{"current_steps": 158, "total_steps": 292, "loss": 0.032, "lr": 4.356259112737096e-06, "epoch": 1.0821917808219177, "percentage": 54.11, "elapsed_time": "0:06:55", "remaining_time": "0:05:52"}
+{"current_steps": 159, "total_steps": 292, "loss": 0.0281, "lr": 4.302950717957304e-06, "epoch": 1.0890410958904109, "percentage": 54.45, "elapsed_time": "0:06:58", "remaining_time": "0:05:50"}
+{"current_steps": 160, "total_steps": 292, "loss": 0.0263, "lr": 4.249723008276737e-06, "epoch": 1.095890410958904, "percentage": 54.79, "elapsed_time": "0:07:00", "remaining_time": "0:05:47"}
+{"current_steps": 161, "total_steps": 292, "loss": 0.0326, "lr": 4.196582144928398e-06, "epoch": 1.1027397260273972, "percentage": 55.14, "elapsed_time": "0:07:03", "remaining_time": "0:05:44"}
+{"current_steps": 162, "total_steps": 292, "loss": 0.0292, "lr": 4.143534279092613e-06, "epoch": 1.1095890410958904, "percentage": 55.48, "elapsed_time": "0:07:05", "remaining_time": "0:05:41"}
+{"current_steps": 163, "total_steps": 292, "loss": 0.0408, "lr": 4.090585551185031e-06, "epoch": 1.1164383561643836, "percentage": 55.82, "elapsed_time": "0:07:08", "remaining_time": "0:05:39"}
+{"current_steps": 164, "total_steps": 292, "loss": 0.0261, "lr": 4.037742090145851e-06, "epoch": 1.1232876712328768, "percentage": 56.16, "elapsed_time": "0:07:10", "remaining_time": "0:05:36"}
+{"current_steps": 165, "total_steps": 292, "loss": 0.023, "lr": 3.985010012730382e-06, "epoch": 1.13013698630137, "percentage": 56.51, "elapsed_time": "0:07:13", "remaining_time": "0:05:33"}
+{"current_steps": 166, "total_steps": 292, "loss": 0.0289, "lr": 3.93239542280102e-06, "epoch": 1.1369863013698631, "percentage": 56.85, "elapsed_time": "0:07:16", "remaining_time": "0:05:31"}
+{"current_steps": 167, "total_steps": 292, "loss": 0.0254, "lr": 3.879904410620703e-06, "epoch": 1.143835616438356, "percentage": 57.19, "elapsed_time": "0:07:18", "remaining_time": "0:05:28"}
+{"current_steps": 168, "total_steps": 292, "loss": 0.0401, "lr": 3.827543052147952e-06, "epoch": 1.1506849315068493, "percentage": 57.53, "elapsed_time": "0:07:21", "remaining_time": "0:05:25"}
+{"current_steps": 169, "total_steps": 292, "loss": 0.041, "lr": 3.775317408333571e-06, "epoch": 1.1575342465753424, "percentage": 57.88, "elapsed_time": "0:07:24", "remaining_time": "0:05:23"}
+{"current_steps": 170, "total_steps": 292, "loss": 0.0304, "lr": 3.7232335244190656e-06, "epoch": 1.1643835616438356, "percentage": 58.22, "elapsed_time": "0:07:27", "remaining_time": "0:05:20"}
+{"current_steps": 171, "total_steps": 292, "loss": 0.026, "lr": 3.6712974292369035e-06, "epoch": 1.1712328767123288, "percentage": 58.56, "elapsed_time": "0:07:29", "remaining_time": "0:05:18"}
+{"current_steps": 172, "total_steps": 292, "loss": 0.0248, "lr": 3.6195151345126556e-06, "epoch": 1.178082191780822, "percentage": 58.9, "elapsed_time": "0:07:32", "remaining_time": "0:05:15"}
+{"current_steps": 173, "total_steps": 292, "loss": 0.0327, "lr": 3.5678926341691283e-06, "epoch": 1.1849315068493151, "percentage": 59.25, "elapsed_time": "0:07:34", "remaining_time": "0:05:12"}
+{"current_steps": 174, "total_steps": 292, "loss": 0.0255, "lr": 3.5164359036325483e-06, "epoch": 1.191780821917808, "percentage": 59.59, "elapsed_time": "0:07:37", "remaining_time": "0:05:10"}
+{"current_steps": 175, "total_steps": 292, "loss": 0.0273, "lr": 3.4651508991409016e-06, "epoch": 1.1986301369863013, "percentage": 59.93, "elapsed_time": "0:07:39", "remaining_time": "0:05:07"}
+{"current_steps": 176, "total_steps": 292, "loss": 0.0334, "lr": 3.4140435570544708e-06, "epoch": 1.2054794520547945, "percentage": 60.27, "elapsed_time": "0:07:42", "remaining_time": "0:05:04"}
+{"current_steps": 177, "total_steps": 292, "loss": 0.0258, "lr": 3.363119793168704e-06, "epoch": 1.2123287671232876, "percentage": 60.62, "elapsed_time": "0:07:45", "remaining_time": "0:05:02"}
+{"current_steps": 178, "total_steps": 292, "loss": 0.0336, "lr": 3.3123855020294344e-06, "epoch": 1.2191780821917808, "percentage": 60.96, "elapsed_time": "0:07:48", "remaining_time": "0:04:59"}
+{"current_steps": 179, "total_steps": 292, "loss": 0.0241, "lr": 3.26184655625058e-06, "epoch": 1.226027397260274, "percentage": 61.3, "elapsed_time": "0:07:50", "remaining_time": "0:04:57"}
+{"current_steps": 180, "total_steps": 292, "loss": 0.0198, "lr": 3.2115088058343725e-06, "epoch": 1.2328767123287672, "percentage": 61.64, "elapsed_time": "0:07:53", "remaining_time": "0:04:54"}
+{"current_steps": 181, "total_steps": 292, "loss": 0.0203, "lr": 3.161378077494205e-06, "epoch": 1.2397260273972603, "percentage": 61.99, "elapsed_time": "0:07:55", "remaining_time": "0:04:51"}
+{"current_steps": 182, "total_steps": 292, "loss": 0.0247, "lr": 3.111460173980175e-06, "epoch": 1.2465753424657535, "percentage": 62.33, "elapsed_time": "0:07:58", "remaining_time": "0:04:49"}
+{"current_steps": 183, "total_steps": 292, "loss": 0.0212, "lr": 3.06176087340741e-06, "epoch": 1.2534246575342465, "percentage": 62.67, "elapsed_time": "0:08:01", "remaining_time": "0:04:46"}
+{"current_steps": 184, "total_steps": 292, "loss": 0.028, "lr": 3.0122859285872214e-06, "epoch": 1.2602739726027397, "percentage": 63.01, "elapsed_time": "0:08:03", "remaining_time": "0:04:44"}
+{"current_steps": 185, "total_steps": 292, "loss": 0.0272, "lr": 2.9630410663612226e-06, "epoch": 1.2671232876712328, "percentage": 63.36, "elapsed_time": "0:08:06", "remaining_time": "0:04:41"}
+{"current_steps": 186, "total_steps": 292, "loss": 0.0301, "lr": 2.914031986938417e-06, "epoch": 1.273972602739726, "percentage": 63.7, "elapsed_time": "0:08:09", "remaining_time": "0:04:38"}
+{"current_steps": 187, "total_steps": 292, "loss": 0.028, "lr": 2.865264363235396e-06, "epoch": 1.2808219178082192, "percentage": 64.04, "elapsed_time": "0:08:11", "remaining_time": "0:04:36"}
+{"current_steps": 188, "total_steps": 292, "loss": 0.0287, "lr": 2.816743840219681e-06, "epoch": 1.2876712328767124, "percentage": 64.38, "elapsed_time": "0:08:14", "remaining_time": "0:04:33"}
+{"current_steps": 189, "total_steps": 292, "loss": 0.0266, "lr": 2.7684760342563045e-06, "epoch": 1.2945205479452055, "percentage": 64.73, "elapsed_time": "0:08:17", "remaining_time": "0:04:30"}
+{"current_steps": 190, "total_steps": 292, "loss": 0.0221, "lr": 2.720466532457707e-06, "epoch": 1.3013698630136985, "percentage": 65.07, "elapsed_time": "0:08:19", "remaining_time": "0:04:28"}
+{"current_steps": 191, "total_steps": 292, "loss": 0.0284, "lr": 2.6727208920370063e-06, "epoch": 1.308219178082192, "percentage": 65.41, "elapsed_time": "0:08:22", "remaining_time": "0:04:25"}
+{"current_steps": 192, "total_steps": 292, "loss": 0.026, "lr": 2.6252446396647503e-06, "epoch": 1.3150684931506849, "percentage": 65.75, "elapsed_time": "0:08:24", "remaining_time": "0:04:22"}
+{"current_steps": 193, "total_steps": 292, "loss": 0.0246, "lr": 2.578043270829178e-06, "epoch": 1.321917808219178, "percentage": 66.1, "elapsed_time": "0:08:27", "remaining_time": "0:04:20"}
+{"current_steps": 194, "total_steps": 292, "loss": 0.03, "lr": 2.531122249200114e-06, "epoch": 1.3287671232876712, "percentage": 66.44, "elapsed_time": "0:08:29", "remaining_time": "0:04:17"}
+{"current_steps": 195, "total_steps": 292, "loss": 0.0262, "lr": 2.4844870059965337e-06, "epoch": 1.3356164383561644, "percentage": 66.78, "elapsed_time": "0:08:32", "remaining_time": "0:04:14"}
+{"current_steps": 196, "total_steps": 292, "loss": 0.0307, "lr": 2.438142939357882e-06, "epoch": 1.3424657534246576, "percentage": 67.12, "elapsed_time": "0:08:34", "remaining_time": "0:04:12"}
+{"current_steps": 197, "total_steps": 292, "loss": 0.0316, "lr": 2.392095413719231e-06, "epoch": 1.3493150684931507, "percentage": 67.47, "elapsed_time": "0:08:37", "remaining_time": "0:04:09"}
+{"current_steps": 198, "total_steps": 292, "loss": 0.0326, "lr": 2.346349759190332e-06, "epoch": 1.356164383561644, "percentage": 67.81, "elapsed_time": "0:08:40", "remaining_time": "0:04:06"}
+{"current_steps": 199, "total_steps": 292, "loss": 0.0287, "lr": 2.3009112709386454e-06, "epoch": 1.3630136986301369, "percentage": 68.15, "elapsed_time": "0:08:43", "remaining_time": "0:04:04"}
+{"current_steps": 200, "total_steps": 292, "loss": 0.0233, "lr": 2.2557852085764053e-06, "epoch": 1.36986301369863, "percentage": 68.49, "elapsed_time": "0:08:45", "remaining_time": "0:04:01"}
+{"current_steps": 200, "total_steps": 292, "eval_loss": 0.12782636284828186, "epoch": 1.36986301369863, "percentage": 68.49, "elapsed_time": "0:08:49", "remaining_time": "0:04:03"}
+{"current_steps": 201, "total_steps": 292, "loss": 0.0205, "lr": 2.2109767955518135e-06, "epoch": 1.3767123287671232, "percentage": 68.84, "elapsed_time": "0:08:51", "remaining_time": "0:04:00"}
+{"current_steps": 202, "total_steps": 292, "loss": 0.0234, "lr": 2.1664912185444127e-06, "epoch": 1.3835616438356164, "percentage": 69.18, "elapsed_time": "0:08:54", "remaining_time": "0:03:58"}
+{"current_steps": 203, "total_steps": 292, "loss": 0.0249, "lr": 2.1223336268647154e-06, "epoch": 1.3904109589041096, "percentage": 69.52, "elapsed_time": "0:08:57", "remaining_time": "0:03:55"}
+{"current_steps": 204, "total_steps": 292, "loss": 0.0224, "lr": 2.0785091318581577e-06, "epoch": 1.3972602739726028, "percentage": 69.86, "elapsed_time": "0:08:59", "remaining_time": "0:03:52"}
+{"current_steps": 205, "total_steps": 292, "loss": 0.0302, "lr": 2.035022806313449e-06, "epoch": 1.404109589041096, "percentage": 70.21, "elapsed_time": "0:09:02", "remaining_time": "0:03:50"}
+{"current_steps": 206, "total_steps": 292, "loss": 0.0273, "lr": 1.991879683875386e-06, "epoch": 1.410958904109589, "percentage": 70.55, "elapsed_time": "0:09:04", "remaining_time": "0:03:47"}
+{"current_steps": 207, "total_steps": 292, "loss": 0.022, "lr": 1.9490847584621993e-06, "epoch": 1.4178082191780823, "percentage": 70.89, "elapsed_time": "0:09:07", "remaining_time": "0:03:44"}
+{"current_steps": 208, "total_steps": 292, "loss": 0.0216, "lr": 1.9066429836874844e-06, "epoch": 1.4246575342465753, "percentage": 71.23, "elapsed_time": "0:09:09", "remaining_time": "0:03:42"}
+{"current_steps": 209, "total_steps": 292, "loss": 0.032, "lr": 1.8645592722868223e-06, "epoch": 1.4315068493150684, "percentage": 71.58, "elapsed_time": "0:09:12", "remaining_time": "0:03:39"}
+{"current_steps": 210, "total_steps": 292, "loss": 0.0301, "lr": 1.8228384955491136e-06, "epoch": 1.4383561643835616, "percentage": 71.92, "elapsed_time": "0:09:14", "remaining_time": "0:03:36"}
+{"current_steps": 211, "total_steps": 292, "loss": 0.0236, "lr": 1.7814854827527144e-06, "epoch": 1.4452054794520548, "percentage": 72.26, "elapsed_time": "0:09:17", "remaining_time": "0:03:34"}
+{"current_steps": 212, "total_steps": 292, "loss": 0.0309, "lr": 1.7405050206064372e-06, "epoch": 1.452054794520548, "percentage": 72.6, "elapsed_time": "0:09:19", "remaining_time": "0:03:31"}
+{"current_steps": 213, "total_steps": 292, "loss": 0.0187, "lr": 1.6999018526954775e-06, "epoch": 1.4589041095890412, "percentage": 72.95, "elapsed_time": "0:09:23", "remaining_time": "0:03:28"}
+{"current_steps": 214, "total_steps": 292, "loss": 0.0241, "lr": 1.6596806789323317e-06, "epoch": 1.4657534246575343, "percentage": 73.29, "elapsed_time": "0:09:25", "remaining_time": "0:03:26"}
+{"current_steps": 215, "total_steps": 292, "loss": 0.0239, "lr": 1.6198461550127758e-06, "epoch": 1.4726027397260273, "percentage": 73.63, "elapsed_time": "0:09:28", "remaining_time": "0:03:23"}
+{"current_steps": 216, "total_steps": 292, "loss": 0.0231, "lr": 1.5804028918769488e-06, "epoch": 1.4794520547945205, "percentage": 73.97, "elapsed_time": "0:09:31", "remaining_time": "0:03:21"}
+{"current_steps": 217, "total_steps": 292, "loss": 0.0282, "lr": 1.5413554551756321e-06, "epoch": 1.4863013698630136, "percentage": 74.32, "elapsed_time": "0:09:33", "remaining_time": "0:03:18"}
+{"current_steps": 218, "total_steps": 292, "loss": 0.0292, "lr": 1.5027083647417657e-06, "epoch": 1.4931506849315068, "percentage": 74.66, "elapsed_time": "0:09:36", "remaining_time": "0:03:15"}
+{"current_steps": 219, "total_steps": 292, "loss": 0.0277, "lr": 1.4644660940672628e-06, "epoch": 1.5, "percentage": 75.0, "elapsed_time": "0:09:38", "remaining_time": "0:03:12"}
+{"current_steps": 220, "total_steps": 292, "loss": 0.0232, "lr": 1.4266330697851955e-06, "epoch": 1.5068493150684932, "percentage": 75.34, "elapsed_time": "0:09:41", "remaining_time": "0:03:10"}
+{"current_steps": 221, "total_steps": 292, "loss": 0.0187, "lr": 1.3892136711573983e-06, "epoch": 1.5136986301369864, "percentage": 75.68, "elapsed_time": "0:09:44", "remaining_time": "0:03:07"}
+{"current_steps": 222, "total_steps": 292, "loss": 0.0228, "lr": 1.3522122295675616e-06, "epoch": 1.5205479452054793, "percentage": 76.03, "elapsed_time": "0:09:46", "remaining_time": "0:03:05"}
+{"current_steps": 223, "total_steps": 292, "loss": 0.0344, "lr": 1.3156330280198637e-06, "epoch": 1.5273972602739727, "percentage": 76.37, "elapsed_time": "0:09:49", "remaining_time": "0:03:02"}
+{"current_steps": 224, "total_steps": 292, "loss": 0.0258, "lr": 1.2794803006431984e-06, "epoch": 1.5342465753424657, "percentage": 76.71, "elapsed_time": "0:09:51", "remaining_time": "0:02:59"}
+{"current_steps": 225, "total_steps": 292, "loss": 0.0229, "lr": 1.2437582322010672e-06, "epoch": 1.541095890410959, "percentage": 77.05, "elapsed_time": "0:09:54", "remaining_time": "0:02:56"}
+{"current_steps": 226, "total_steps": 292, "loss": 0.0306, "lr": 1.2084709576071885e-06, "epoch": 1.547945205479452, "percentage": 77.4, "elapsed_time": "0:09:56", "remaining_time": "0:02:54"}
+{"current_steps": 227, "total_steps": 292, "loss": 0.0195, "lr": 1.1736225614468627e-06, "epoch": 1.5547945205479452, "percentage": 77.74, "elapsed_time": "0:09:59", "remaining_time": "0:02:51"}
+{"current_steps": 228, "total_steps": 292, "loss": 0.0356, "lr": 1.1392170775041788e-06, "epoch": 1.5616438356164384, "percentage": 78.08, "elapsed_time": "0:10:02", "remaining_time": "0:02:48"}
+{"current_steps": 229, "total_steps": 292, "loss": 0.0265, "lr": 1.1052584882950896e-06, "epoch": 1.5684931506849316, "percentage": 78.42, "elapsed_time": "0:10:04", "remaining_time": "0:02:46"}
+{"current_steps": 230, "total_steps": 292, "loss": 0.0264, "lr": 1.0717507246064273e-06, "epoch": 1.5753424657534247, "percentage": 78.77, "elapsed_time": "0:10:07", "remaining_time": "0:02:43"}
+{"current_steps": 231, "total_steps": 292, "loss": 0.0216, "lr": 1.0386976650409102e-06, "epoch": 1.5821917808219177, "percentage": 79.11, "elapsed_time": "0:10:10", "remaining_time": "0:02:41"}
+{"current_steps": 232, "total_steps": 292, "loss": 0.0243, "lr": 1.0061031355681766e-06, "epoch": 1.589041095890411, "percentage": 79.45, "elapsed_time": "0:10:12", "remaining_time": "0:02:38"}
+{"current_steps": 233, "total_steps": 292, "loss": 0.0267, "lr": 9.739709090819254e-07, "epoch": 1.595890410958904, "percentage": 79.79, "elapsed_time": "0:10:14", "remaining_time": "0:02:35"}
+{"current_steps": 234, "total_steps": 292, "loss": 0.0216, "lr": 9.423047049631956e-07, "epoch": 1.6027397260273972, "percentage": 80.14, "elapsed_time": "0:10:18", "remaining_time": "0:02:33"}
+{"current_steps": 235, "total_steps": 292, "loss": 0.0224, "lr": 9.111081886498374e-07, "epoch": 1.6095890410958904, "percentage": 80.48, "elapsed_time": "0:10:20", "remaining_time": "0:02:30"}
+{"current_steps": 236, "total_steps": 292, "loss": 0.0189, "lr": 8.803849712122292e-07, "epoch": 1.6164383561643836, "percentage": 80.82, "elapsed_time": "0:10:23", "remaining_time": "0:02:27"}
+{"current_steps": 237, "total_steps": 292, "loss": 0.025, "lr": 8.501386089352858e-07, "epoch": 1.6232876712328768, "percentage": 81.16, "elapsed_time": "0:10:26", "remaining_time": "0:02:25"}
+{"current_steps": 238, "total_steps": 292, "loss": 0.0276, "lr": 8.203726029068149e-07, "epoch": 1.6301369863013697, "percentage": 81.51, "elapsed_time": "0:10:28", "remaining_time": "0:02:22"}
+{"current_steps": 239, "total_steps": 292, "loss": 0.0219, "lr": 7.910903986122537e-07, "epoch": 1.6369863013698631, "percentage": 81.85, "elapsed_time": "0:10:31", "remaining_time": "0:02:19"}
+{"current_steps": 240, "total_steps": 292, "loss": 0.0246, "lr": 7.622953855358456e-07, "epoch": 1.643835616438356, "percentage": 82.19, "elapsed_time": "0:10:33", "remaining_time": "0:02:17"}
+{"current_steps": 241, "total_steps": 292, "loss": 0.0266, "lr": 7.339908967683007e-07, "epoch": 1.6506849315068495, "percentage": 82.53, "elapsed_time": "0:10:36", "remaining_time": "0:02:14"}
+{"current_steps": 242, "total_steps": 292, "loss": 0.0231, "lr": 7.061802086209857e-07, "epoch": 1.6575342465753424, "percentage": 82.88, "elapsed_time": "0:10:38", "remaining_time": "0:02:11"}
+{"current_steps": 243, "total_steps": 292, "loss": 0.0234, "lr": 6.788665402466782e-07, "epoch": 1.6643835616438356, "percentage": 83.22, "elapsed_time": "0:10:41", "remaining_time": "0:02:09"}
+{"current_steps": 244, "total_steps": 292, "loss": 0.0228, "lr": 6.52053053266945e-07, "epoch": 1.6712328767123288, "percentage": 83.56, "elapsed_time": "0:10:44", "remaining_time": "0:02:06"}
+{"current_steps": 245, "total_steps": 292, "loss": 0.0157, "lr": 6.257428514061764e-07, "epoch": 1.678082191780822, "percentage": 83.9, "elapsed_time": "0:10:47", "remaining_time": "0:02:04"}
+{"current_steps": 246, "total_steps": 292, "loss": 0.0223, "lr": 5.999389801323219e-07, "epoch": 1.6849315068493151, "percentage": 84.25, "elapsed_time": "0:10:49", "remaining_time": "0:02:01"}
+{"current_steps": 247, "total_steps": 292, "loss": 0.0265, "lr": 5.746444263043715e-07, "epoch": 1.691780821917808, "percentage": 84.59, "elapsed_time": "0:10:52", "remaining_time": "0:01:58"}
+{"current_steps": 248, "total_steps": 292, "loss": 0.0213, "lr": 5.498621178266167e-07, "epoch": 1.6986301369863015, "percentage": 84.93, "elapsed_time": "0:10:54", "remaining_time": "0:01:56"}
+{"current_steps": 249, "total_steps": 292, "loss": 0.0238, "lr": 5.255949233097451e-07, "epoch": 1.7054794520547945, "percentage": 85.27, "elapsed_time": "0:10:57", "remaining_time": "0:01:53"}
+{"current_steps": 250, "total_steps": 292, "loss": 0.0254, "lr": 5.018456517387837e-07, "epoch": 1.7123287671232876, "percentage": 85.62, "elapsed_time": "0:11:00", "remaining_time": "0:01:50"}
+{"current_steps": 251, "total_steps": 292, "loss": 0.0269, "lr": 4.786170521479588e-07, "epoch": 1.7191780821917808, "percentage": 85.96, "elapsed_time": "0:11:02", "remaining_time": "0:01:48"}
+{"current_steps": 252, "total_steps": 292, "loss": 0.0337, "lr": 4.5591181330248534e-07, "epoch": 1.726027397260274, "percentage": 86.3, "elapsed_time": "0:11:05", "remaining_time": "0:01:45"}
+{"current_steps": 253, "total_steps": 292, "loss": 0.0184, "lr": 4.3373256338733847e-07, "epoch": 1.7328767123287672, "percentage": 86.64, "elapsed_time": "0:11:07", "remaining_time": "0:01:42"}
+{"current_steps": 254, "total_steps": 292, "loss": 0.0274, "lr": 4.1208186970303097e-07, "epoch": 1.7397260273972601, "percentage": 86.99, "elapsed_time": "0:11:10", "remaining_time": "0:01:40"}
+{"current_steps": 255, "total_steps": 292, "loss": 0.0281, "lr": 3.90962238368448e-07, "epoch": 1.7465753424657535, "percentage": 87.33, "elapsed_time": "0:11:13", "remaining_time": "0:01:37"}
+{"current_steps": 256, "total_steps": 292, "loss": 0.02, "lr": 3.70376114030751e-07, "epoch": 1.7534246575342465, "percentage": 87.67, "elapsed_time": "0:11:15", "remaining_time": "0:01:35"}
+{"current_steps": 257, "total_steps": 292, "loss": 0.0207, "lr": 3.503258795824105e-07, "epoch": 1.7602739726027399, "percentage": 88.01, "elapsed_time": "0:11:18", "remaining_time": "0:01:32"}
+{"current_steps": 258, "total_steps": 292, "loss": 0.0249, "lr": 3.308138558853746e-07, "epoch": 1.7671232876712328, "percentage": 88.36, "elapsed_time": "0:11:20", "remaining_time": "0:01:29"}
+{"current_steps": 259, "total_steps": 292, "loss": 0.0221, "lr": 3.1184230150243025e-07, "epoch": 1.773972602739726, "percentage": 88.7, "elapsed_time": "0:11:23", "remaining_time": "0:01:27"}
+{"current_steps": 260, "total_steps": 292, "loss": 0.0231, "lr": 2.934134124357646e-07, "epoch": 1.7808219178082192, "percentage": 89.04, "elapsed_time": "0:11:25", "remaining_time": "0:01:24"}
+{"current_steps": 261, "total_steps": 292, "loss": 0.0247, "lr": 2.755293218727739e-07, "epoch": 1.7876712328767124, "percentage": 89.38, "elapsed_time": "0:11:28", "remaining_time": "0:01:21"}
+{"current_steps": 262, "total_steps": 292, "loss": 0.0254, "lr": 2.5819209993914185e-07, "epoch": 1.7945205479452055, "percentage": 89.73, "elapsed_time": "0:11:30", "remaining_time": "0:01:19"}
+{"current_steps": 263, "total_steps": 292, "loss": 0.0182, "lr": 2.4140375345921895e-07, "epoch": 1.8013698630136985, "percentage": 90.07, "elapsed_time": "0:11:33", "remaining_time": "0:01:16"}
+{"current_steps": 264, "total_steps": 292, "loss": 0.0213, "lr": 2.2516622572372416e-07, "epoch": 1.808219178082192, "percentage": 90.41, "elapsed_time": "0:11:36", "remaining_time": "0:01:13"}
+{"current_steps": 265, "total_steps": 292, "loss": 0.0286, "lr": 2.094813962648101e-07, "epoch": 1.8150684931506849, "percentage": 90.75, "elapsed_time": "0:11:38", "remaining_time": "0:01:11"}
+{"current_steps": 266, "total_steps": 292, "loss": 0.0199, "lr": 1.9435108063849684e-07, "epoch": 1.821917808219178, "percentage": 91.1, "elapsed_time": "0:11:41", "remaining_time": "0:01:08"}
+{"current_steps": 267, "total_steps": 292, "loss": 0.0282, "lr": 1.7977703021452185e-07, "epoch": 1.8287671232876712, "percentage": 91.44, "elapsed_time": "0:11:43", "remaining_time": "0:01:05"}
+{"current_steps": 268, "total_steps": 292, "loss": 0.0243, "lr": 1.6576093197361253e-07, "epoch": 1.8356164383561644, "percentage": 91.78, "elapsed_time": "0:11:46", "remaining_time": "0:01:03"}
+{"current_steps": 269, "total_steps": 292, "loss": 0.0244, "lr": 1.523044083122138e-07, "epoch": 1.8424657534246576, "percentage": 92.12, "elapsed_time": "0:11:49", "remaining_time": "0:01:00"}
+{"current_steps": 270, "total_steps": 292, "loss": 0.028, "lr": 1.39409016854693e-07, "epoch": 1.8493150684931505, "percentage": 92.47, "elapsed_time": "0:11:51", "remaining_time": "0:00:57"}
+{"current_steps": 271, "total_steps": 292, "loss": 0.0252, "lr": 1.2707625027304104e-07, "epoch": 1.856164383561644, "percentage": 92.81, "elapsed_time": "0:11:54", "remaining_time": "0:00:55"}
+{"current_steps": 272, "total_steps": 292, "loss": 0.0268, "lr": 1.1530753611409151e-07, "epoch": 1.8630136986301369, "percentage": 93.15, "elapsed_time": "0:11:56", "remaining_time": "0:00:52"}
+{"current_steps": 273, "total_steps": 292, "loss": 0.0256, "lr": 1.041042366342787e-07, "epoch": 1.8698630136986303, "percentage": 93.49, "elapsed_time": "0:11:59", "remaining_time": "0:00:50"}
+{"current_steps": 274, "total_steps": 292, "loss": 0.0256, "lr": 9.346764864195335e-08, "epoch": 1.8767123287671232, "percentage": 93.84, "elapsed_time": "0:12:01", "remaining_time": "0:00:47"}
+{"current_steps": 275, "total_steps": 292, "loss": 0.0315, "lr": 8.339900334727536e-08, "epoch": 1.8835616438356164, "percentage": 94.18, "elapsed_time": "0:12:04", "remaining_time": "0:00:44"}
+{"current_steps": 276, "total_steps": 292, "loss": 0.0225, "lr": 7.389946621969679e-08, "epoch": 1.8904109589041096, "percentage": 94.52, "elapsed_time": "0:12:07", "remaining_time": "0:00:42"}
+{"current_steps": 277, "total_steps": 292, "loss": 0.0205, "lr": 6.497013685305586e-08, "epoch": 1.8972602739726028, "percentage": 94.86, "elapsed_time": "0:12:10", "remaining_time": "0:00:39"}
+{"current_steps": 278, "total_steps": 292, "loss": 0.0219, "lr": 5.661204883829763e-08, "epoch": 1.904109589041096, "percentage": 95.21, "elapsed_time": "0:12:12", "remaining_time": "0:00:36"}
+{"current_steps": 279, "total_steps": 292, "loss": 0.0269, "lr": 4.8826169643832464e-08, "epoch": 1.910958904109589, "percentage": 95.55, "elapsed_time": "0:12:15", "remaining_time": "0:00:34"}
+{"current_steps": 280, "total_steps": 292, "loss": 0.0267, "lr": 4.1613400503550114e-08, "epoch": 1.9178082191780823, "percentage": 95.89, "elapsed_time": "0:12:18", "remaining_time": "0:00:31"}
+{"current_steps": 281, "total_steps": 292, "loss": 0.0272, "lr": 3.4974576312497564e-08, "epoch": 1.9246575342465753, "percentage": 96.23, "elapsed_time": "0:12:21", "remaining_time": "0:00:29"}
+{"current_steps": 282, "total_steps": 292, "loss": 0.021, "lr": 2.8910465530240793e-08, "epoch": 1.9315068493150684, "percentage": 96.58, "elapsed_time": "0:12:23", "remaining_time": "0:00:26"}
+{"current_steps": 283, "total_steps": 292, "loss": 0.0312, "lr": 2.3421770091912044e-08, "epoch": 1.9383561643835616, "percentage": 96.92, "elapsed_time": "0:12:26", "remaining_time": "0:00:23"}
+{"current_steps": 284, "total_steps": 292, "loss": 0.0268, "lr": 1.850912532696092e-08, "epoch": 1.9452054794520548, "percentage": 97.26, "elapsed_time": "0:12:28", "remaining_time": "0:00:21"}
+{"current_steps": 285, "total_steps": 292, "loss": 0.0244, "lr": 1.4173099885610997e-08, "epoch": 1.952054794520548, "percentage": 97.6, "elapsed_time": "0:12:31", "remaining_time": "0:00:18"}
+{"current_steps": 286, "total_steps": 292, "loss": 0.0258, "lr": 1.041419567303914e-08, "epoch": 1.958904109589041, "percentage": 97.95, "elapsed_time": "0:12:35", "remaining_time": "0:00:15"}
+{"current_steps": 287, "total_steps": 292, "loss": 0.0228, "lr": 7.2328477912769756e-09, "epoch": 1.9657534246575343, "percentage": 98.29, "elapsed_time": "0:12:37", "remaining_time": "0:00:13"}
+{"current_steps": 288, "total_steps": 292, "loss": 0.0213, "lr": 4.629424488850065e-09, "epoch": 1.9726027397260273, "percentage": 98.63, "elapsed_time": "0:12:40", "remaining_time": "0:00:10"}
+{"current_steps": 289, "total_steps": 292, "loss": 0.0232, "lr": 2.604227118148117e-09, "epoch": 1.9794520547945207, "percentage": 98.97, "elapsed_time": "0:12:42", "remaining_time": "0:00:07"}
+{"current_steps": 290, "total_steps": 292, "loss": 0.0224, "lr": 1.1574901005456662e-09, "epoch": 1.9863013698630136, "percentage": 99.32, "elapsed_time": "0:12:45", "remaining_time": "0:00:05"}
+{"current_steps": 291, "total_steps": 292, "loss": 0.0254, "lr": 2.89380899267111e-10, "epoch": 1.9931506849315068, "percentage": 99.66, "elapsed_time": "0:12:47", "remaining_time": "0:00:02"}
+{"current_steps": 292, "total_steps": 292, "loss": 0.0212, "lr": 0.0, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "0:12:50", "remaining_time": "0:00:00"}
+{"current_steps": 292, "total_steps": 292, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "0:13:53", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,2094 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 200,
+  "global_step": 292,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00684931506849315,
+      "grad_norm": 0.9593834756973793,
+      "learning_rate": 9.999710619100732e-06,
+      "loss": 0.1231,
+      "step": 1
+    },
+    {
+      "epoch": 0.0136986301369863,
+      "grad_norm": 1.0366318450053902,
+      "learning_rate": 9.998842509899456e-06,
+      "loss": 0.1599,
+      "step": 2
+    },
+    {
+      "epoch": 0.02054794520547945,
+      "grad_norm": 0.6508585223819798,
+      "learning_rate": 9.997395772881853e-06,
+      "loss": 0.1042,
+      "step": 3
+    },
+    {
+      "epoch": 0.0273972602739726,
+      "grad_norm": 0.7657469202490956,
+      "learning_rate": 9.995370575511151e-06,
+      "loss": 0.114,
+      "step": 4
+    },
+    {
+      "epoch": 0.03424657534246575,
+      "grad_norm": 0.773450378447972,
+      "learning_rate": 9.992767152208724e-06,
+      "loss": 0.0993,
+      "step": 5
+    },
+    {
+      "epoch": 0.0410958904109589,
+      "grad_norm": 0.6264523234725071,
+      "learning_rate": 9.989585804326963e-06,
+      "loss": 0.0887,
+      "step": 6
+    },
+    {
+      "epoch": 0.04794520547945205,
+      "grad_norm": 0.7537995289017994,
+      "learning_rate": 9.985826900114391e-06,
+      "loss": 0.0992,
+      "step": 7
+    },
+    {
+      "epoch": 0.0547945205479452,
+      "grad_norm": 0.6902981451332041,
+      "learning_rate": 9.98149087467304e-06,
+      "loss": 0.0936,
+      "step": 8
+    },
+    {
+      "epoch": 0.06164383561643835,
+      "grad_norm": 0.7066295696196615,
+      "learning_rate": 9.97657822990809e-06,
+      "loss": 0.1073,
+      "step": 9
+    },
+    {
+      "epoch": 0.0684931506849315,
+      "grad_norm": 0.7056273280149921,
+      "learning_rate": 9.97108953446976e-06,
+      "loss": 0.0993,
+      "step": 10
+    },
+    {
+      "epoch": 0.07534246575342465,
+      "grad_norm": 0.8400214910429082,
+      "learning_rate": 9.965025423687505e-06,
+      "loss": 0.1084,
+      "step": 11
+    },
+    {
+      "epoch": 0.0821917808219178,
+      "grad_norm": 0.6548937347185184,
+      "learning_rate": 9.95838659949645e-06,
+      "loss": 0.0843,
+      "step": 12
+    },
+    {
+      "epoch": 0.08904109589041095,
+      "grad_norm": 0.6559408616855686,
+      "learning_rate": 9.951173830356168e-06,
+      "loss": 0.0894,
+      "step": 13
+    },
+    {
+      "epoch": 0.0958904109589041,
+      "grad_norm": 0.8356138855690183,
+      "learning_rate": 9.943387951161702e-06,
+      "loss": 0.1161,
+      "step": 14
+    },
+    {
+      "epoch": 0.10273972602739725,
+      "grad_norm": 0.6727072374629645,
+      "learning_rate": 9.935029863146946e-06,
+      "loss": 0.0872,
+      "step": 15
+    },
+    {
+      "epoch": 0.1095890410958904,
+      "grad_norm": 0.6205940232841655,
+      "learning_rate": 9.926100533780304e-06,
+      "loss": 0.0697,
+      "step": 16
+    },
+    {
+      "epoch": 0.11643835616438356,
+      "grad_norm": 0.8038247362280305,
+      "learning_rate": 9.916600996652726e-06,
+      "loss": 0.1001,
+      "step": 17
+    },
+    {
+      "epoch": 0.1232876712328767,
+      "grad_norm": 0.8703467157302192,
+      "learning_rate": 9.906532351358047e-06,
+      "loss": 0.1142,
+      "step": 18
+    },
+    {
+      "epoch": 0.13013698630136986,
+      "grad_norm": 0.749160596257599,
+      "learning_rate": 9.895895763365722e-06,
+      "loss": 0.0951,
+      "step": 19
+    },
+    {
+      "epoch": 0.136986301369863,
+      "grad_norm": 0.7754609595123414,
+      "learning_rate": 9.88469246388591e-06,
+      "loss": 0.1102,
+      "step": 20
+    },
+    {
+      "epoch": 0.14383561643835616,
+      "grad_norm": 0.7830298124590415,
+      "learning_rate": 9.872923749726959e-06,
+      "loss": 0.1102,
+      "step": 21
+    },
+    {
+      "epoch": 0.1506849315068493,
+      "grad_norm": 0.7078241134141917,
+      "learning_rate": 9.860590983145307e-06,
+      "loss": 0.0889,
+      "step": 22
+    },
+    {
+      "epoch": 0.15753424657534246,
+      "grad_norm": 0.8428672025857301,
+      "learning_rate": 9.847695591687788e-06,
+      "loss": 0.1128,
+      "step": 23
+    },
+    {
+      "epoch": 0.1643835616438356,
+      "grad_norm": 0.7220487174797594,
+      "learning_rate": 9.834239068026388e-06,
+      "loss": 0.099,
+      "step": 24
+    },
+    {
+      "epoch": 0.17123287671232876,
+      "grad_norm": 0.7064317613231897,
+      "learning_rate": 9.82022296978548e-06,
+      "loss": 0.0931,
+      "step": 25
+    },
+    {
+      "epoch": 0.1780821917808219,
+      "grad_norm": 0.6914687462455392,
+      "learning_rate": 9.805648919361505e-06,
+      "loss": 0.0875,
+      "step": 26
+    },
+    {
+      "epoch": 0.18493150684931506,
+      "grad_norm": 0.7424459388678591,
+      "learning_rate": 9.790518603735191e-06,
+      "loss": 0.1036,
+      "step": 27
+    },
+    {
+      "epoch": 0.1917808219178082,
+      "grad_norm": 0.735070163197544,
+      "learning_rate": 9.774833774276278e-06,
+      "loss": 0.0994,
+      "step": 28
+    },
+    {
+      "epoch": 0.19863013698630136,
+      "grad_norm": 0.8280959253160824,
+      "learning_rate": 9.758596246540782e-06,
+      "loss": 0.1192,
+      "step": 29
+    },
+    {
+      "epoch": 0.2054794520547945,
+      "grad_norm": 0.6597645775525501,
+      "learning_rate": 9.741807900060858e-06,
+      "loss": 0.0928,
+      "step": 30
+    },
+    {
+      "epoch": 0.21232876712328766,
+      "grad_norm": 0.763238425774405,
+      "learning_rate": 9.724470678127226e-06,
+      "loss": 0.1115,
+      "step": 31
+    },
+    {
+      "epoch": 0.2191780821917808,
+      "grad_norm": 0.704884344301177,
+      "learning_rate": 9.706586587564236e-06,
+      "loss": 0.109,
+      "step": 32
+    },
+    {
+      "epoch": 0.22602739726027396,
+      "grad_norm": 0.7005139949025764,
+      "learning_rate": 9.68815769849757e-06,
+      "loss": 0.1025,
+      "step": 33
+    },
+    {
+      "epoch": 0.2328767123287671,
+      "grad_norm": 0.805148188033095,
+      "learning_rate": 9.669186144114627e-06,
+      "loss": 0.1096,
+      "step": 34
+    },
+    {
+      "epoch": 0.23972602739726026,
+      "grad_norm": 0.7624408584402959,
+      "learning_rate": 9.649674120417591e-06,
+      "loss": 0.1092,
+      "step": 35
+    },
+    {
+      "epoch": 0.2465753424657534,
+      "grad_norm": 0.8131922614568783,
+      "learning_rate": 9.62962388596925e-06,
+      "loss": 0.1106,
+      "step": 36
+    },
+    {
+      "epoch": 0.2534246575342466,
+      "grad_norm": 0.8147548648435898,
+      "learning_rate": 9.609037761631552e-06,
+      "loss": 0.1164,
+      "step": 37
+    },
+    {
+      "epoch": 0.2602739726027397,
+      "grad_norm": 0.7389791528033364,
+      "learning_rate": 9.587918130296969e-06,
+      "loss": 0.1041,
+      "step": 38
+    },
+    {
+      "epoch": 0.2671232876712329,
+      "grad_norm": 0.7059065795605491,
+      "learning_rate": 9.566267436612662e-06,
+      "loss": 0.0961,
+      "step": 39
+    },
+    {
+      "epoch": 0.273972602739726,
+      "grad_norm": 0.6760733340665249,
+      "learning_rate": 9.544088186697515e-06,
+      "loss": 0.1033,
+      "step": 40
+    },
+    {
+      "epoch": 0.2808219178082192,
+      "grad_norm": 0.5673291242923014,
+      "learning_rate": 9.521382947852042e-06,
+      "loss": 0.0692,
+      "step": 41
+    },
+    {
+      "epoch": 0.2876712328767123,
+      "grad_norm": 0.7574437710944075,
+      "learning_rate": 9.498154348261217e-06,
+      "loss": 0.1024,
+      "step": 42
+    },
+    {
+      "epoch": 0.2945205479452055,
+      "grad_norm": 0.7041195553676743,
+      "learning_rate": 9.474405076690257e-06,
+      "loss": 0.1045,
+      "step": 43
+    },
+    {
+      "epoch": 0.3013698630136986,
+      "grad_norm": 0.6677385477116528,
+      "learning_rate": 9.450137882173385e-06,
+      "loss": 0.088,
+      "step": 44
+    },
+    {
+      "epoch": 0.3082191780821918,
+      "grad_norm": 0.6346381036971247,
+      "learning_rate": 9.425355573695628e-06,
+      "loss": 0.089,
+      "step": 45
+    },
+    {
+      "epoch": 0.3150684931506849,
+      "grad_norm": 0.8287038911978735,
+      "learning_rate": 9.40006101986768e-06,
+      "loss": 0.1076,
+      "step": 46
+    },
+    {
+      "epoch": 0.3219178082191781,
+      "grad_norm": 0.6982754394550266,
+      "learning_rate": 9.374257148593824e-06,
+      "loss": 0.0997,
+      "step": 47
+    },
+    {
+      "epoch": 0.3287671232876712,
+      "grad_norm": 0.7973278910995713,
+      "learning_rate": 9.347946946733055e-06,
+      "loss": 0.1089,
+      "step": 48
+    },
+    {
+      "epoch": 0.3356164383561644,
+      "grad_norm": 0.7330982797613164,
+      "learning_rate": 9.321133459753322e-06,
+      "loss": 0.1028,
+      "step": 49
+    },
+    {
+      "epoch": 0.3424657534246575,
+      "grad_norm": 0.7760454288407581,
+      "learning_rate": 9.293819791379016e-06,
+      "loss": 0.1166,
+      "step": 50
+    },
+    {
+      "epoch": 0.3493150684931507,
+      "grad_norm": 0.7600040395504281,
+      "learning_rate": 9.266009103231702e-06,
+      "loss": 0.1008,
+      "step": 51
+    },
+    {
+      "epoch": 0.3561643835616438,
+      "grad_norm": 0.8462820099684166,
+      "learning_rate": 9.237704614464157e-06,
+      "loss": 0.1126,
+      "step": 52
+    },
+    {
+      "epoch": 0.363013698630137,
+      "grad_norm": 0.7809644134536166,
+      "learning_rate": 9.208909601387748e-06,
+      "loss": 0.1081,
+      "step": 53
+    },
+    {
+      "epoch": 0.3698630136986301,
+      "grad_norm": 0.6806410480801305,
+      "learning_rate": 9.179627397093184e-06,
+      "loss": 0.0901,
+      "step": 54
+    },
+    {
+      "epoch": 0.3767123287671233,
+      "grad_norm": 0.6999141874681865,
+      "learning_rate": 9.149861391064714e-06,
+      "loss": 0.1017,
+      "step": 55
+    },
+    {
+      "epoch": 0.3835616438356164,
+      "grad_norm": 0.7362016061187906,
+      "learning_rate": 9.119615028787771e-06,
+      "loss": 0.1018,
+      "step": 56
+    },
+    {
+      "epoch": 0.3904109589041096,
+      "grad_norm": 0.796603408754819,
+      "learning_rate": 9.088891811350164e-06,
+      "loss": 0.1049,
+      "step": 57
+    },
+    {
+      "epoch": 0.3972602739726027,
+      "grad_norm": 0.7576527967226971,
+      "learning_rate": 9.057695295036806e-06,
+      "loss": 0.1042,
+      "step": 58
+    },
+    {
+      "epoch": 0.4041095890410959,
+      "grad_norm": 0.6427964831166068,
+      "learning_rate": 9.026029090918076e-06,
+      "loss": 0.0888,
+      "step": 59
+    },
+    {
+      "epoch": 0.410958904109589,
+      "grad_norm": 0.7481130379993173,
+      "learning_rate": 8.993896864431825e-06,
+      "loss": 0.1002,
+      "step": 60
+    },
+    {
+      "epoch": 0.4178082191780822,
+      "grad_norm": 0.6574786390576889,
+      "learning_rate": 8.96130233495909e-06,
+      "loss": 0.0881,
+      "step": 61
+    },
+    {
+      "epoch": 0.4246575342465753,
+      "grad_norm": 0.6754357749197657,
+      "learning_rate": 8.928249275393572e-06,
+      "loss": 0.0944,
+      "step": 62
+    },
+    {
+      "epoch": 0.4315068493150685,
+      "grad_norm": 0.6509678298973031,
+      "learning_rate": 8.894741511704911e-06,
+      "loss": 0.0874,
+      "step": 63
+    },
+    {
+      "epoch": 0.4383561643835616,
+      "grad_norm": 0.7125323018469738,
+      "learning_rate": 8.860782922495821e-06,
+      "loss": 0.1053,
+      "step": 64
+    },
+    {
+      "epoch": 0.4452054794520548,
+      "grad_norm": 0.7898553497859033,
+      "learning_rate": 8.826377438553138e-06,
+      "loss": 0.1098,
+      "step": 65
+    },
+    {
+      "epoch": 0.4520547945205479,
+      "grad_norm": 0.7270237572097952,
+      "learning_rate": 8.791529042392813e-06,
+      "loss": 0.0983,
+      "step": 66
+    },
+    {
+      "epoch": 0.4589041095890411,
+      "grad_norm": 0.6556709322157142,
+      "learning_rate": 8.756241767798934e-06,
+      "loss": 0.0922,
+      "step": 67
+    },
+    {
+      "epoch": 0.4657534246575342,
+      "grad_norm": 0.7330312516853001,
+      "learning_rate": 8.720519699356804e-06,
+      "loss": 0.1009,
+      "step": 68
+    },
+    {
+      "epoch": 0.4726027397260274,
+      "grad_norm": 0.7894192501819968,
+      "learning_rate": 8.684366971980139e-06,
+      "loss": 0.1136,
+      "step": 69
+    },
+    {
+      "epoch": 0.4794520547945205,
+      "grad_norm": 0.7620445217510202,
+      "learning_rate": 8.647787770432439e-06,
+      "loss": 0.106,
+      "step": 70
+    },
+    {
+      "epoch": 0.4863013698630137,
+      "grad_norm": 0.6309422495845969,
+      "learning_rate": 8.610786328842602e-06,
+      "loss": 0.0781,
+      "step": 71
+    },
+    {
+      "epoch": 0.4931506849315068,
+      "grad_norm": 0.7065600537534685,
+      "learning_rate": 8.573366930214807e-06,
+      "loss": 0.0953,
+      "step": 72
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.7115736455959387,
+      "learning_rate": 8.535533905932739e-06,
+      "loss": 0.0977,
+      "step": 73
+    },
+    {
+      "epoch": 0.5068493150684932,
+      "grad_norm": 0.6999241584756221,
+      "learning_rate": 8.497291635258235e-06,
+      "loss": 0.0986,
+      "step": 74
+    },
+    {
+      "epoch": 0.5136986301369864,
+      "grad_norm": 0.8052347118427793,
+      "learning_rate": 8.458644544824371e-06,
+      "loss": 0.1066,
+      "step": 75
+    },
+    {
+      "epoch": 0.5205479452054794,
+      "grad_norm": 0.5684986456446331,
+      "learning_rate": 8.419597108123054e-06,
+      "loss": 0.0778,
+      "step": 76
+    },
+    {
+      "epoch": 0.5273972602739726,
+      "grad_norm": 0.6569153273607293,
+      "learning_rate": 8.380153844987225e-06,
+      "loss": 0.0885,
+      "step": 77
+    },
+    {
+      "epoch": 0.5342465753424658,
+      "grad_norm": 0.6543237226637773,
+      "learning_rate": 8.340319321067668e-06,
+      "loss": 0.0784,
+      "step": 78
+    },
+    {
+      "epoch": 0.541095890410959,
+      "grad_norm": 0.7091685827973174,
+      "learning_rate": 8.300098147304523e-06,
+      "loss": 0.1051,
+      "step": 79
+    },
+    {
+      "epoch": 0.547945205479452,
+      "grad_norm": 0.9118299925604293,
+      "learning_rate": 8.259494979393563e-06,
+      "loss": 0.1233,
+      "step": 80
+    },
+    {
+      "epoch": 0.5547945205479452,
+      "grad_norm": 0.6996397199985479,
+      "learning_rate": 8.218514517247287e-06,
+      "loss": 0.1009,
+      "step": 81
+    },
+    {
+      "epoch": 0.5616438356164384,
+      "grad_norm": 0.7319616286987115,
+      "learning_rate": 8.177161504450887e-06,
+      "loss": 0.0909,
+      "step": 82
+    },
+    {
+      "epoch": 0.5684931506849316,
+      "grad_norm": 0.7321313442715763,
+      "learning_rate": 8.135440727713179e-06,
+      "loss": 0.0965,
+      "step": 83
+    },
+    {
+      "epoch": 0.5753424657534246,
+      "grad_norm": 0.8922413248256611,
+      "learning_rate": 8.093357016312518e-06,
+      "loss": 0.1382,
+      "step": 84
+    },
+    {
+      "epoch": 0.5821917808219178,
+      "grad_norm": 0.8165751985420481,
+      "learning_rate": 8.050915241537802e-06,
+      "loss": 0.111,
+      "step": 85
+    },
+    {
+      "epoch": 0.589041095890411,
+      "grad_norm": 0.646721720523638,
+      "learning_rate": 8.008120316124612e-06,
+      "loss": 0.0903,
+      "step": 86
+    },
+    {
+      "epoch": 0.5958904109589042,
+      "grad_norm": 0.6836085431096287,
+      "learning_rate": 7.964977193686551e-06,
+      "loss": 0.0889,
+      "step": 87
+    },
+    {
+      "epoch": 0.6027397260273972,
+      "grad_norm": 0.7903980806529376,
+      "learning_rate": 7.921490868141843e-06,
+      "loss": 0.1025,
+      "step": 88
+    },
+    {
+      "epoch": 0.6095890410958904,
+      "grad_norm": 0.6568790448248424,
+      "learning_rate": 7.877666373135287e-06,
+      "loss": 0.0872,
+      "step": 89
+    },
+    {
+      "epoch": 0.6164383561643836,
+      "grad_norm": 0.7609820250058389,
+      "learning_rate": 7.83350878145559e-06,
+      "loss": 0.0976,
+      "step": 90
+    },
+    {
+      "epoch": 0.6232876712328768,
+      "grad_norm": 0.6136717465608816,
+      "learning_rate": 7.789023204448189e-06,
+      "loss": 0.0829,
+      "step": 91
+    },
+    {
+      "epoch": 0.6301369863013698,
+      "grad_norm": 0.6912288388438279,
+      "learning_rate": 7.744214791423597e-06,
+      "loss": 0.0862,
+      "step": 92
+    },
+    {
+      "epoch": 0.636986301369863,
+      "grad_norm": 0.8508509328110958,
+      "learning_rate": 7.699088729061355e-06,
+      "loss": 0.1227,
+      "step": 93
+    },
+    {
+      "epoch": 0.6438356164383562,
+      "grad_norm": 0.6878684979870703,
+      "learning_rate": 7.653650240809667e-06,
+      "loss": 0.0939,
+      "step": 94
+    },
+    {
+      "epoch": 0.6506849315068494,
+      "grad_norm": 0.7319689001661777,
+      "learning_rate": 7.60790458628077e-06,
+      "loss": 0.0957,
+      "step": 95
+    },
+    {
+      "epoch": 0.6575342465753424,
+      "grad_norm": 0.7472566372429058,
+      "learning_rate": 7.56185706064212e-06,
+      "loss": 0.0975,
+      "step": 96
+    },
+    {
+      "epoch": 0.6643835616438356,
+      "grad_norm": 0.6527988826637163,
+      "learning_rate": 7.5155129940034675e-06,
+      "loss": 0.0857,
+      "step": 97
+    },
+    {
+      "epoch": 0.6712328767123288,
+      "grad_norm": 0.7451882863373878,
+      "learning_rate": 7.468877750799887e-06,
+      "loss": 0.0995,
+      "step": 98
+    },
+    {
+      "epoch": 0.678082191780822,
+      "grad_norm": 0.6169267993802472,
+      "learning_rate": 7.421956729170823e-06,
+      "loss": 0.0779,
+      "step": 99
+    },
+    {
+      "epoch": 0.684931506849315,
+      "grad_norm": 0.6927338328840201,
+      "learning_rate": 7.374755360335253e-06,
+      "loss": 0.0908,
+      "step": 100
+    },
+    {
+      "epoch": 0.6917808219178082,
+      "grad_norm": 0.7461895323803881,
+      "learning_rate": 7.327279107962995e-06,
+      "loss": 0.094,
+      "step": 101
+    },
+    {
+      "epoch": 0.6986301369863014,
+      "grad_norm": 0.7678847568540728,
+      "learning_rate": 7.279533467542295e-06,
+      "loss": 0.1122,
+      "step": 102
+    },
+    {
+      "epoch": 0.7054794520547946,
+      "grad_norm": 0.747944270140034,
+      "learning_rate": 7.2315239657436955e-06,
+      "loss": 0.094,
+      "step": 103
+    },
+    {
+      "epoch": 0.7123287671232876,
+      "grad_norm": 0.8445562397107534,
+      "learning_rate": 7.183256159780321e-06,
+      "loss": 0.1022,
+      "step": 104
+    },
+    {
+      "epoch": 0.7191780821917808,
+      "grad_norm": 0.7483595593058716,
+      "learning_rate": 7.134735636764606e-06,
+      "loss": 0.0998,
+      "step": 105
+    },
+    {
+      "epoch": 0.726027397260274,
+      "grad_norm": 0.7776383870936369,
+      "learning_rate": 7.085968013061585e-06,
+      "loss": 0.1122,
+      "step": 106
+    },
+    {
+      "epoch": 0.7328767123287672,
+      "grad_norm": 0.6336882385097793,
+      "learning_rate": 7.036958933638779e-06,
+      "loss": 0.074,
+      "step": 107
+    },
+    {
+      "epoch": 0.7397260273972602,
+      "grad_norm": 0.8050211422597002,
+      "learning_rate": 6.987714071412781e-06,
+      "loss": 0.1103,
+      "step": 108
+    },
+    {
+      "epoch": 0.7465753424657534,
+      "grad_norm": 0.767920393441073,
+      "learning_rate": 6.938239126592592e-06,
+      "loss": 0.1014,
+      "step": 109
+    },
+    {
+      "epoch": 0.7534246575342466,
+      "grad_norm": 0.7736203031736444,
+      "learning_rate": 6.888539826019824e-06,
+      "loss": 0.0908,
+      "step": 110
+    },
+    {
+      "epoch": 0.7602739726027398,
+      "grad_norm": 0.7336376174406862,
+      "learning_rate": 6.8386219225057945e-06,
+      "loss": 0.0973,
+      "step": 111
+    },
+    {
+      "epoch": 0.7671232876712328,
+      "grad_norm": 0.7394637744781307,
+      "learning_rate": 6.788491194165629e-06,
+      "loss": 0.0983,
+      "step": 112
+    },
+    {
+      "epoch": 0.773972602739726,
+      "grad_norm": 0.657050477816448,
+      "learning_rate": 6.738153443749421e-06,
+      "loss": 0.101,
+      "step": 113
+    },
+    {
+      "epoch": 0.7808219178082192,
+      "grad_norm": 0.7903607521289585,
+      "learning_rate": 6.687614497970567e-06,
+      "loss": 0.0949,
+      "step": 114
+    },
+    {
+      "epoch": 0.7876712328767124,
+      "grad_norm": 0.742230209437882,
+      "learning_rate": 6.636880206831298e-06,
+      "loss": 0.1011,
+      "step": 115
+    },
+    {
+      "epoch": 0.7945205479452054,
+      "grad_norm": 0.6961952816223166,
+      "learning_rate": 6.585956442945531e-06,
+      "loss": 0.0872,
+      "step": 116
+    },
+    {
+      "epoch": 0.8013698630136986,
+      "grad_norm": 0.6312585843346639,
+      "learning_rate": 6.534849100859101e-06,
+      "loss": 0.0859,
+      "step": 117
+    },
+    {
+      "epoch": 0.8082191780821918,
+      "grad_norm": 0.7692870373605378,
+      "learning_rate": 6.483564096367452e-06,
+      "loss": 0.1142,
+      "step": 118
+    },
+    {
+      "epoch": 0.815068493150685,
+      "grad_norm": 0.6797050781586332,
+      "learning_rate": 6.432107365830872e-06,
+      "loss": 0.1047,
+      "step": 119
+    },
+    {
+      "epoch": 0.821917808219178,
+      "grad_norm": 0.7264370261717285,
+      "learning_rate": 6.380484865487346e-06,
+      "loss": 0.1174,
+      "step": 120
+    },
+    {
+      "epoch": 0.8287671232876712,
+      "grad_norm": 0.7138217749375094,
+      "learning_rate": 6.328702570763098e-06,
+      "loss": 0.1014,
+      "step": 121
+    },
+    {
+      "epoch": 0.8356164383561644,
+      "grad_norm": 0.6864241460283704,
+      "learning_rate": 6.276766475580935e-06,
+      "loss": 0.0888,
+      "step": 122
+    },
+    {
+      "epoch": 0.8424657534246576,
+      "grad_norm": 0.7861601718762463,
+      "learning_rate": 6.224682591666431e-06,
+      "loss": 0.1194,
+      "step": 123
+    },
+    {
+      "epoch": 0.8493150684931506,
+      "grad_norm": 0.6308802102336228,
+      "learning_rate": 6.1724569478520495e-06,
+      "loss": 0.0766,
+      "step": 124
+    },
+    {
+      "epoch": 0.8561643835616438,
+      "grad_norm": 0.7463619271167676,
+      "learning_rate": 6.120095589379299e-06,
+      "loss": 0.0999,
+      "step": 125
+    },
+    {
+      "epoch": 0.863013698630137,
+      "grad_norm": 0.723089306950006,
+      "learning_rate": 6.067604577198981e-06,
+      "loss": 0.1026,
+      "step": 126
+    },
+    {
+      "epoch": 0.8698630136986302,
+      "grad_norm": 0.8128427236232567,
+      "learning_rate": 6.014989987269617e-06,
+      "loss": 0.1177,
+      "step": 127
+    },
+    {
+      "epoch": 0.8767123287671232,
+      "grad_norm": 0.6557388016832486,
+      "learning_rate": 5.96225790985415e-06,
+      "loss": 0.0841,
+      "step": 128
+    },
+    {
+      "epoch": 0.8835616438356164,
+      "grad_norm": 0.5731463675088502,
+      "learning_rate": 5.909414448814971e-06,
+      "loss": 0.0668,
+      "step": 129
+    },
+    {
+      "epoch": 0.8904109589041096,
+      "grad_norm": 0.827794610936805,
+      "learning_rate": 5.856465720907388e-06,
+      "loss": 0.1175,
+      "step": 130
+    },
+    {
+      "epoch": 0.8972602739726028,
+      "grad_norm": 0.7900302418716827,
+      "learning_rate": 5.803417855071603e-06,
+      "loss": 0.1083,
+      "step": 131
+    },
+    {
+      "epoch": 0.9041095890410958,
+      "grad_norm": 0.6967481596548395,
+      "learning_rate": 5.7502769917232635e-06,
+      "loss": 0.1048,
+      "step": 132
+    },
+    {
+      "epoch": 0.910958904109589,
+      "grad_norm": 0.7431142462170286,
+      "learning_rate": 5.6970492820426994e-06,
+      "loss": 0.0984,
+      "step": 133
+    },
+    {
+      "epoch": 0.9178082191780822,
+      "grad_norm": 0.706859004886188,
+      "learning_rate": 5.643740887262905e-06,
+      "loss": 0.0909,
+      "step": 134
+    },
+    {
+      "epoch": 0.9246575342465754,
+      "grad_norm": 0.7439038100978435,
+      "learning_rate": 5.59035797795637e-06,
+      "loss": 0.1065,
+      "step": 135
+    },
+    {
+      "epoch": 0.9315068493150684,
+      "grad_norm": 0.7336777940517607,
+      "learning_rate": 5.536906733320816e-06,
+      "loss": 0.0923,
+      "step": 136
+    },
+    {
+      "epoch": 0.9383561643835616,
+      "grad_norm": 0.6486216523289474,
+      "learning_rate": 5.483393340463938e-06,
+      "loss": 0.0903,
+      "step": 137
+    },
+    {
+      "epoch": 0.9452054794520548,
+      "grad_norm": 0.738475458968018,
+      "learning_rate": 5.429823993687234e-06,
+      "loss": 0.1001,
+      "step": 138
+    },
+    {
+      "epoch": 0.952054794520548,
+      "grad_norm": 0.6195239351601184,
+      "learning_rate": 5.376204893769e-06,
+      "loss": 0.0828,
+      "step": 139
+    },
+    {
+      "epoch": 0.958904109589041,
+      "grad_norm": 0.6597631011909114,
+      "learning_rate": 5.322542247246583e-06,
+      "loss": 0.0895,
+      "step": 140
+    },
+    {
+      "epoch": 0.9657534246575342,
+      "grad_norm": 0.6247250694020478,
+      "learning_rate": 5.26884226569794e-06,
+      "loss": 0.0877,
+      "step": 141
+    },
+    {
+      "epoch": 0.9726027397260274,
+      "grad_norm": 0.6907307523255068,
+      "learning_rate": 5.215111165022653e-06,
+      "loss": 0.0935,
+      "step": 142
+    },
+    {
+      "epoch": 0.9794520547945206,
+      "grad_norm": 0.7157610963876196,
+      "learning_rate": 5.161355164722416e-06,
+      "loss": 0.105,
+      "step": 143
+    },
+    {
+      "epoch": 0.9863013698630136,
+      "grad_norm": 0.6481915828513748,
+      "learning_rate": 5.107580487181112e-06,
+      "loss": 0.0811,
+      "step": 144
+    },
+    {
+      "epoch": 0.9931506849315068,
+      "grad_norm": 0.5761545491887491,
+      "learning_rate": 5.0537933569445585e-06,
+      "loss": 0.0776,
+      "step": 145
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.6969449252695846,
+      "learning_rate": 5e-06,
+      "loss": 0.0813,
+      "step": 146
+    },
+    {
+      "epoch": 1.0068493150684932,
+      "grad_norm": 0.40022211319597883,
+      "learning_rate": 4.946206643055443e-06,
+      "loss": 0.035,
+      "step": 147
+    },
+    {
+      "epoch": 1.0136986301369864,
+      "grad_norm": 0.36996841036837175,
+      "learning_rate": 4.89241951281889e-06,
+      "loss": 0.0242,
+      "step": 148
+    },
+    {
+      "epoch": 1.0205479452054795,
+      "grad_norm": 0.5889368440525016,
+      "learning_rate": 4.838644835277585e-06,
+      "loss": 0.0393,
+      "step": 149
+    },
+    {
+      "epoch": 1.0273972602739727,
+      "grad_norm": 0.3820739637144992,
+      "learning_rate": 4.784888834977347e-06,
+      "loss": 0.0299,
+      "step": 150
+    },
+    {
+      "epoch": 1.0342465753424657,
+      "grad_norm": 0.5379931997949438,
+      "learning_rate": 4.731157734302063e-06,
+      "loss": 0.0408,
+      "step": 151
+    },
+    {
+      "epoch": 1.0410958904109588,
+      "grad_norm": 0.3693305843552891,
+      "learning_rate": 4.6774577527534195e-06,
+      "loss": 0.0294,
+      "step": 152
+    },
+    {
+      "epoch": 1.047945205479452,
+      "grad_norm": 0.4888264501648255,
+      "learning_rate": 4.623795106231001e-06,
+      "loss": 0.0402,
+      "step": 153
+    },
+    {
+      "epoch": 1.0547945205479452,
+      "grad_norm": 0.4134341070423952,
+      "learning_rate": 4.570176006312769e-06,
+      "loss": 0.0306,
+      "step": 154
+    },
+    {
+      "epoch": 1.0616438356164384,
+      "grad_norm": 0.4186096948787815,
+      "learning_rate": 4.516606659536063e-06,
+      "loss": 0.0325,
+      "step": 155
+    },
+    {
+      "epoch": 1.0684931506849316,
+      "grad_norm": 0.5562678365427425,
+      "learning_rate": 4.463093266679185e-06,
+      "loss": 0.041,
+      "step": 156
+    },
+    {
+      "epoch": 1.0753424657534247,
+      "grad_norm": 0.41023005785548794,
+      "learning_rate": 4.40964202204363e-06,
+      "loss": 0.0278,
+      "step": 157
+    },
+    {
+      "epoch": 1.0821917808219177,
+      "grad_norm": 0.4409264685189105,
+      "learning_rate": 4.356259112737096e-06,
+      "loss": 0.032,
+      "step": 158
+    },
+    {
+      "epoch": 1.0890410958904109,
+      "grad_norm": 0.41264943175896446,
+      "learning_rate": 4.302950717957304e-06,
+      "loss": 0.0281,
+      "step": 159
+    },
+    {
+      "epoch": 1.095890410958904,
+      "grad_norm": 0.42074167821091196,
+      "learning_rate": 4.249723008276737e-06,
+      "loss": 0.0263,
+      "step": 160
+    },
+    {
+      "epoch": 1.1027397260273972,
+      "grad_norm": 0.6080366405585456,
+      "learning_rate": 4.196582144928398e-06,
+      "loss": 0.0326,
+      "step": 161
+    },
+    {
+      "epoch": 1.1095890410958904,
+      "grad_norm": 0.4999009139678011,
+      "learning_rate": 4.143534279092613e-06,
+      "loss": 0.0292,
+      "step": 162
+    },
+    {
+      "epoch": 1.1164383561643836,
+      "grad_norm": 0.6964542936679394,
+      "learning_rate": 4.090585551185031e-06,
+      "loss": 0.0408,
+      "step": 163
+    },
+    {
+      "epoch": 1.1232876712328768,
+      "grad_norm": 0.48487172127635747,
+      "learning_rate": 4.037742090145851e-06,
+      "loss": 0.0261,
+      "step": 164
+    },
+    {
+      "epoch": 1.13013698630137,
+      "grad_norm": 0.410607594047809,
+      "learning_rate": 3.985010012730382e-06,
+      "loss": 0.023,
+      "step": 165
+    },
+    {
+      "epoch": 1.1369863013698631,
+      "grad_norm": 0.488422520515759,
+      "learning_rate": 3.93239542280102e-06,
+      "loss": 0.0289,
+      "step": 166
+    },
+    {
+      "epoch": 1.143835616438356,
+      "grad_norm": 0.47037371049545285,
+      "learning_rate": 3.879904410620703e-06,
+      "loss": 0.0254,
+      "step": 167
+    },
+    {
+      "epoch": 1.1506849315068493,
+      "grad_norm": 0.6996438632030281,
+      "learning_rate": 3.827543052147952e-06,
+      "loss": 0.0401,
+      "step": 168
+    },
+    {
+      "epoch": 1.1575342465753424,
+      "grad_norm": 0.6482935966804099,
+      "learning_rate": 3.775317408333571e-06,
+      "loss": 0.041,
+      "step": 169
+    },
+    {
+      "epoch": 1.1643835616438356,
+      "grad_norm": 0.5348791028923409,
+      "learning_rate": 3.7232335244190656e-06,
+      "loss": 0.0304,
+      "step": 170
+    },
+    {
+      "epoch": 1.1712328767123288,
+      "grad_norm": 0.6824645649985049,
+      "learning_rate": 3.6712974292369035e-06,
+      "loss": 0.026,
+      "step": 171
+    },
+    {
+      "epoch": 1.178082191780822,
+      "grad_norm": 0.5222355032203745,
+      "learning_rate": 3.6195151345126556e-06,
+      "loss": 0.0248,
+      "step": 172
+    },
+    {
+      "epoch": 1.1849315068493151,
+      "grad_norm": 0.6436945077378039,
+      "learning_rate": 3.5678926341691283e-06,
+      "loss": 0.0327,
+      "step": 173
+    },
+    {
+      "epoch": 1.191780821917808,
+      "grad_norm": 0.4815469429272036,
+      "learning_rate": 3.5164359036325483e-06,
+      "loss": 0.0255,
+      "step": 174
+    },
+    {
+      "epoch": 1.1986301369863013,
+      "grad_norm": 0.6105258583091369,
+      "learning_rate": 3.4651508991409016e-06,
+      "loss": 0.0273,
+      "step": 175
+    },
+    {
+      "epoch": 1.2054794520547945,
+      "grad_norm": 0.7436330215160468,
+      "learning_rate": 3.4140435570544708e-06,
+      "loss": 0.0334,
+      "step": 176
+    },
+    {
+      "epoch": 1.2123287671232876,
+      "grad_norm": 0.5039349276354249,
+      "learning_rate": 3.363119793168704e-06,
+      "loss": 0.0258,
+      "step": 177
+    },
+    {
+      "epoch": 1.2191780821917808,
+      "grad_norm": 0.7800878927795044,
+      "learning_rate": 3.3123855020294344e-06,
+      "loss": 0.0336,
+      "step": 178
+    },
+    {
+      "epoch": 1.226027397260274,
+      "grad_norm": 0.5432005292406065,
+      "learning_rate": 3.26184655625058e-06,
+      "loss": 0.0241,
+      "step": 179
+    },
+    {
+      "epoch": 1.2328767123287672,
+      "grad_norm": 0.4788982798162723,
+      "learning_rate": 3.2115088058343725e-06,
+      "loss": 0.0198,
+      "step": 180
+    },
+    {
+      "epoch": 1.2397260273972603,
+      "grad_norm": 0.5332736570342096,
+      "learning_rate": 3.161378077494205e-06,
+      "loss": 0.0203,
+      "step": 181
+    },
+    {
+      "epoch": 1.2465753424657535,
+      "grad_norm": 0.6018244302290219,
+      "learning_rate": 3.111460173980175e-06,
+      "loss": 0.0247,
+      "step": 182
+    },
+    {
+      "epoch": 1.2534246575342465,
+      "grad_norm": 0.4714750435429761,
+      "learning_rate": 3.06176087340741e-06,
+      "loss": 0.0212,
+      "step": 183
+    },
+    {
+      "epoch": 1.2602739726027397,
+      "grad_norm": 0.5173165747302255,
+      "learning_rate": 3.0122859285872214e-06,
+      "loss": 0.028,
+      "step": 184
+    },
+    {
+      "epoch": 1.2671232876712328,
+      "grad_norm": 0.5340467637611864,
+      "learning_rate": 2.9630410663612226e-06,
+      "loss": 0.0272,
+      "step": 185
+    },
+    {
+      "epoch": 1.273972602739726,
+      "grad_norm": 0.6775945861048723,
+      "learning_rate": 2.914031986938417e-06,
+      "loss": 0.0301,
+      "step": 186
+    },
+    {
+      "epoch": 1.2808219178082192,
+      "grad_norm": 0.5833141967005896,
+      "learning_rate": 2.865264363235396e-06,
+      "loss": 0.028,
+      "step": 187
+    },
+    {
+      "epoch": 1.2876712328767124,
+      "grad_norm": 0.655533441118794,
+      "learning_rate": 2.816743840219681e-06,
+      "loss": 0.0287,
+      "step": 188
+    },
+    {
+      "epoch": 1.2945205479452055,
+      "grad_norm": 0.5511262540225013,
+      "learning_rate": 2.7684760342563045e-06,
+      "loss": 0.0266,
+      "step": 189
+    },
+    {
+      "epoch": 1.3013698630136985,
+      "grad_norm": 0.5252392637950674,
+      "learning_rate": 2.720466532457707e-06,
+      "loss": 0.0221,
+      "step": 190
+    },
+    {
+      "epoch": 1.308219178082192,
+      "grad_norm": 0.5688735522785462,
+      "learning_rate": 2.6727208920370063e-06,
+      "loss": 0.0284,
+      "step": 191
+    },
+    {
+      "epoch": 1.3150684931506849,
+      "grad_norm": 0.6107186649801015,
+      "learning_rate": 2.6252446396647503e-06,
+      "loss": 0.026,
+      "step": 192
+    },
+    {
+      "epoch": 1.321917808219178,
+      "grad_norm": 0.5508581227820072,
+      "learning_rate": 2.578043270829178e-06,
+      "loss": 0.0246,
+      "step": 193
+    },
+    {
+      "epoch": 1.3287671232876712,
+      "grad_norm": 0.6823030158545308,
+      "learning_rate": 2.531122249200114e-06,
+      "loss": 0.03,
+      "step": 194
+    },
+    {
+      "epoch": 1.3356164383561644,
+      "grad_norm": 0.5426691220715485,
+      "learning_rate": 2.4844870059965337e-06,
+      "loss": 0.0262,
+      "step": 195
+    },
+    {
+      "epoch": 1.3424657534246576,
+      "grad_norm": 0.6755955651401638,
+      "learning_rate": 2.438142939357882e-06,
+      "loss": 0.0307,
+      "step": 196
+    },
+    {
+      "epoch": 1.3493150684931507,
+      "grad_norm": 0.6100256813331606,
+      "learning_rate": 2.392095413719231e-06,
+      "loss": 0.0316,
+      "step": 197
+    },
+    {
+      "epoch": 1.356164383561644,
+      "grad_norm": 0.6463879601128194,
+      "learning_rate": 2.346349759190332e-06,
+      "loss": 0.0326,
+      "step": 198
+    },
+    {
+      "epoch": 1.3630136986301369,
+      "grad_norm": 0.5778357649758314,
+      "learning_rate": 2.3009112709386454e-06,
+      "loss": 0.0287,
+      "step": 199
+    },
+    {
+      "epoch": 1.36986301369863,
+      "grad_norm": 0.5411990616960265,
+      "learning_rate": 2.2557852085764053e-06,
+      "loss": 0.0233,
+      "step": 200
+    },
+    {
+      "epoch": 1.36986301369863,
+      "eval_loss": 0.12782636284828186,
+      "eval_runtime": 3.606,
+      "eval_samples_per_second": 6.656,
+      "eval_steps_per_second": 1.664,
+      "step": 200
+    },
+    {
+      "epoch": 1.3767123287671232,
+      "grad_norm": 0.4606747783778257,
+      "learning_rate": 2.2109767955518135e-06,
+      "loss": 0.0205,
+      "step": 201
+    },
+    {
+      "epoch": 1.3835616438356164,
+      "grad_norm": 0.5644346220855605,
+      "learning_rate": 2.1664912185444127e-06,
+      "loss": 0.0234,
+      "step": 202
+    },
+    {
+      "epoch": 1.3904109589041096,
+      "grad_norm": 0.48298285290916276,
+      "learning_rate": 2.1223336268647154e-06,
+      "loss": 0.0249,
+      "step": 203
+    },
+    {
+      "epoch": 1.3972602739726028,
+      "grad_norm": 0.5415449063607428,
+      "learning_rate": 2.0785091318581577e-06,
+      "loss": 0.0224,
+      "step": 204
+    },
+    {
+      "epoch": 1.404109589041096,
+      "grad_norm": 0.6133699712229087,
+      "learning_rate": 2.035022806313449e-06,
+      "loss": 0.0302,
+      "step": 205
+    },
+    {
+      "epoch": 1.410958904109589,
+      "grad_norm": 0.6514447134449943,
+      "learning_rate": 1.991879683875386e-06,
+      "loss": 0.0273,
+      "step": 206
+    },
+    {
+      "epoch": 1.4178082191780823,
+      "grad_norm": 0.5197921872912981,
+      "learning_rate": 1.9490847584621993e-06,
+      "loss": 0.022,
+      "step": 207
+    },
+    {
+      "epoch": 1.4246575342465753,
+      "grad_norm": 0.48059243417518727,
+      "learning_rate": 1.9066429836874844e-06,
+      "loss": 0.0216,
+      "step": 208
+    },
+    {
+      "epoch": 1.4315068493150684,
+      "grad_norm": 0.6832678607875455,
+      "learning_rate": 1.8645592722868223e-06,
+      "loss": 0.032,
+      "step": 209
+    },
+    {
+      "epoch": 1.4383561643835616,
+      "grad_norm": 0.573234705352323,
+      "learning_rate": 1.8228384955491136e-06,
+      "loss": 0.0301,
+      "step": 210
+    },
+    {
+      "epoch": 1.4452054794520548,
+      "grad_norm": 0.5953906384483378,
+      "learning_rate": 1.7814854827527144e-06,
+      "loss": 0.0236,
+      "step": 211
+    },
+    {
+      "epoch": 1.452054794520548,
+      "grad_norm": 0.6950223099104754,
+      "learning_rate": 1.7405050206064372e-06,
+      "loss": 0.0309,
+      "step": 212
+    },
+    {
+      "epoch": 1.4589041095890412,
+      "grad_norm": 0.5081896383994274,
+      "learning_rate": 1.6999018526954775e-06,
+      "loss": 0.0187,
+      "step": 213
+    },
+    {
+      "epoch": 1.4657534246575343,
+      "grad_norm": 0.5249858681016251,
+      "learning_rate": 1.6596806789323317e-06,
+      "loss": 0.0241,
+      "step": 214
+    },
+    {
+      "epoch": 1.4726027397260273,
+      "grad_norm": 0.5572741587536835,
+      "learning_rate": 1.6198461550127758e-06,
+      "loss": 0.0239,
+      "step": 215
+    },
+    {
+      "epoch": 1.4794520547945205,
+      "grad_norm": 0.529648002049523,
+      "learning_rate": 1.5804028918769488e-06,
+      "loss": 0.0231,
+      "step": 216
+    },
+    {
+      "epoch": 1.4863013698630136,
+      "grad_norm": 0.5699037866710787,
+      "learning_rate": 1.5413554551756321e-06,
+      "loss": 0.0282,
+      "step": 217
+    },
+    {
+      "epoch": 1.4931506849315068,
+      "grad_norm": 0.540327055337863,
+      "learning_rate": 1.5027083647417657e-06,
+      "loss": 0.0292,
+      "step": 218
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.5581385603660535,
+      "learning_rate": 1.4644660940672628e-06,
+      "loss": 0.0277,
+      "step": 219
+    },
+    {
+      "epoch": 1.5068493150684932,
+      "grad_norm": 0.544646773379327,
+      "learning_rate": 1.4266330697851955e-06,
+      "loss": 0.0232,
+      "step": 220
+    },
+    {
+      "epoch": 1.5136986301369864,
+      "grad_norm": 0.4803602047141029,
+      "learning_rate": 1.3892136711573983e-06,
+      "loss": 0.0187,
+      "step": 221
+    },
+    {
+      "epoch": 1.5205479452054793,
+      "grad_norm": 0.5153130049065987,
+      "learning_rate": 1.3522122295675616e-06,
+      "loss": 0.0228,
+      "step": 222
+    },
+    {
+      "epoch": 1.5273972602739727,
+      "grad_norm": 0.6096781687815058,
+      "learning_rate": 1.3156330280198637e-06,
+      "loss": 0.0344,
+      "step": 223
+    },
+    {
+      "epoch": 1.5342465753424657,
+      "grad_norm": 0.6798958042444144,
+      "learning_rate": 1.2794803006431984e-06,
+      "loss": 0.0258,
+      "step": 224
+    },
+    {
+      "epoch": 1.541095890410959,
+      "grad_norm": 0.574437025110474,
+      "learning_rate": 1.2437582322010672e-06,
+      "loss": 0.0229,
+      "step": 225
+    },
+    {
+      "epoch": 1.547945205479452,
+      "grad_norm": 0.581621714813575,
+      "learning_rate": 1.2084709576071885e-06,
+      "loss": 0.0306,
+      "step": 226
+    },
+    {
+      "epoch": 1.5547945205479452,
+      "grad_norm": 0.48322576235961967,
+      "learning_rate": 1.1736225614468627e-06,
+      "loss": 0.0195,
+      "step": 227
+    },
+    {
+      "epoch": 1.5616438356164384,
+      "grad_norm": 0.6228610203390602,
+      "learning_rate": 1.1392170775041788e-06,
+      "loss": 0.0356,
+      "step": 228
+    },
+    {
+      "epoch": 1.5684931506849316,
+      "grad_norm": 0.5106649557751511,
+      "learning_rate": 1.1052584882950896e-06,
+      "loss": 0.0265,
+      "step": 229
+    },
+    {
+      "epoch": 1.5753424657534247,
+      "grad_norm": 0.5699134903899079,
+      "learning_rate": 1.0717507246064273e-06,
+      "loss": 0.0264,
+      "step": 230
+    },
+    {
+      "epoch": 1.5821917808219177,
+      "grad_norm": 0.5082138048285709,
+      "learning_rate": 1.0386976650409102e-06,
+      "loss": 0.0216,
+      "step": 231
+    },
+    {
+      "epoch": 1.589041095890411,
+      "grad_norm": 0.5660787920681069,
+      "learning_rate": 1.0061031355681766e-06,
+      "loss": 0.0243,
+      "step": 232
+    },
+    {
+      "epoch": 1.595890410958904,
+      "grad_norm": 0.564593525018267,
+      "learning_rate": 9.739709090819254e-07,
+      "loss": 0.0267,
+      "step": 233
+    },
+    {
+      "epoch": 1.6027397260273972,
+      "grad_norm": 0.4505995387354749,
+      "learning_rate": 9.423047049631956e-07,
+      "loss": 0.0216,
+      "step": 234
+    },
+    {
+      "epoch": 1.6095890410958904,
+      "grad_norm": 0.49897796704328956,
+      "learning_rate": 9.111081886498374e-07,
+      "loss": 0.0224,
+      "step": 235
+    },
+    {
+      "epoch": 1.6164383561643836,
+      "grad_norm": 0.48246709394966497,
+      "learning_rate": 8.803849712122292e-07,
+      "loss": 0.0189,
+      "step": 236
+    },
+    {
+      "epoch": 1.6232876712328768,
+      "grad_norm": 0.5048987533509625,
+      "learning_rate": 8.501386089352858e-07,
+      "loss": 0.025,
+      "step": 237
+    },
+    {
+      "epoch": 1.6301369863013697,
+      "grad_norm": 0.5303344493323716,
+      "learning_rate": 8.203726029068149e-07,
+      "loss": 0.0276,
+      "step": 238
+    },
+    {
+      "epoch": 1.6369863013698631,
+      "grad_norm": 0.45632405891796,
+      "learning_rate": 7.910903986122537e-07,
+      "loss": 0.0219,
+      "step": 239
+    },
+    {
+      "epoch": 1.643835616438356,
+      "grad_norm": 0.5807701211688502,
+      "learning_rate": 7.622953855358456e-07,
+      "loss": 0.0246,
+      "step": 240
+    },
+    {
+      "epoch": 1.6506849315068495,
+      "grad_norm": 0.5598140370861285,
+      "learning_rate": 7.339908967683007e-07,
+      "loss": 0.0266,
+      "step": 241
+    },
+    {
+      "epoch": 1.6575342465753424,
+      "grad_norm": 0.533124504980895,
+      "learning_rate": 7.061802086209857e-07,
+      "loss": 0.0231,
+      "step": 242
+    },
+    {
+      "epoch": 1.6643835616438356,
+      "grad_norm": 0.487914295898038,
+      "learning_rate": 6.788665402466782e-07,
+      "loss": 0.0234,
+      "step": 243
+    },
+    {
+      "epoch": 1.6712328767123288,
+      "grad_norm": 0.5310859286103036,
+      "learning_rate": 6.52053053266945e-07,
+      "loss": 0.0228,
+      "step": 244
+    },
+    {
+      "epoch": 1.678082191780822,
+      "grad_norm": 0.3943206473013647,
+      "learning_rate": 6.257428514061764e-07,
+      "loss": 0.0157,
+      "step": 245
+    },
+    {
+      "epoch": 1.6849315068493151,
+      "grad_norm": 0.5082623094622303,
+      "learning_rate": 5.999389801323219e-07,
+      "loss": 0.0223,
+      "step": 246
+    },
+    {
+      "epoch": 1.691780821917808,
+      "grad_norm": 0.5520449974096024,
+      "learning_rate": 5.746444263043715e-07,
+      "loss": 0.0265,
+      "step": 247
+    },
+    {
+      "epoch": 1.6986301369863015,
+      "grad_norm": 0.4739165044710361,
+      "learning_rate": 5.498621178266167e-07,
+      "loss": 0.0213,
+      "step": 248
+    },
+    {
+      "epoch": 1.7054794520547945,
+      "grad_norm": 0.543019095966297,
+      "learning_rate": 5.255949233097451e-07,
+      "loss": 0.0238,
+      "step": 249
+    },
+    {
+      "epoch": 1.7123287671232876,
+      "grad_norm": 0.5610967838400929,
+      "learning_rate": 5.018456517387837e-07,
+      "loss": 0.0254,
+      "step": 250
+    },
+    {
+      "epoch": 1.7191780821917808,
+      "grad_norm": 0.6020409159639404,
+      "learning_rate": 4.786170521479588e-07,
+      "loss": 0.0269,
+      "step": 251
+    },
+    {
+      "epoch": 1.726027397260274,
+      "grad_norm": 0.6720413988173921,
+      "learning_rate": 4.5591181330248534e-07,
+      "loss": 0.0337,
+      "step": 252
+    },
+    {
+      "epoch": 1.7328767123287672,
+      "grad_norm": 0.4658420917547858,
+      "learning_rate": 4.3373256338733847e-07,
+      "loss": 0.0184,
+      "step": 253
+    },
+    {
+      "epoch": 1.7397260273972601,
+      "grad_norm": 0.6139558415380756,
+      "learning_rate": 4.1208186970303097e-07,
+      "loss": 0.0274,
+      "step": 254
+    },
+    {
+      "epoch": 1.7465753424657535,
+      "grad_norm": 0.6603667053997885,
+      "learning_rate": 3.90962238368448e-07,
+      "loss": 0.0281,
+      "step": 255
+    },
+    {
+      "epoch": 1.7534246575342465,
+      "grad_norm": 0.533881147566802,
+      "learning_rate": 3.70376114030751e-07,
+      "loss": 0.02,
+      "step": 256
+    },
+    {
+      "epoch": 1.7602739726027399,
+      "grad_norm": 0.5016506115967129,
+      "learning_rate": 3.503258795824105e-07,
+      "loss": 0.0207,
+      "step": 257
+    },
+    {
+      "epoch": 1.7671232876712328,
+      "grad_norm": 0.5933464109254851,
+      "learning_rate": 3.308138558853746e-07,
+      "loss": 0.0249,
+      "step": 258
+    },
+    {
+      "epoch": 1.773972602739726,
+      "grad_norm": 0.5429151978483335,
+      "learning_rate": 3.1184230150243025e-07,
+      "loss": 0.0221,
+      "step": 259
+    },
+    {
+      "epoch": 1.7808219178082192,
+      "grad_norm": 0.5263783415786707,
+      "learning_rate": 2.934134124357646e-07,
+      "loss": 0.0231,
+      "step": 260
+    },
+    {
+      "epoch": 1.7876712328767124,
+      "grad_norm": 0.547290301145945,
+      "learning_rate": 2.755293218727739e-07,
+      "loss": 0.0247,
+      "step": 261
+    },
+    {
+      "epoch": 1.7945205479452055,
+      "grad_norm": 0.5879041613883544,
+      "learning_rate": 2.5819209993914185e-07,
+      "loss": 0.0254,
+      "step": 262
+    },
+    {
+      "epoch": 1.8013698630136985,
+      "grad_norm": 0.4761793429556001,
+      "learning_rate": 2.4140375345921895e-07,
+      "loss": 0.0182,
+      "step": 263
+    },
+    {
+      "epoch": 1.808219178082192,
+      "grad_norm": 0.4464490137111368,
+      "learning_rate": 2.2516622572372416e-07,
+      "loss": 0.0213,
+      "step": 264
+    },
+    {
+      "epoch": 1.8150684931506849,
+      "grad_norm": 0.5484527193426002,
+      "learning_rate": 2.094813962648101e-07,
+      "loss": 0.0286,
+      "step": 265
+    },
+    {
+      "epoch": 1.821917808219178,
+      "grad_norm": 0.4957780798831515,
+      "learning_rate": 1.9435108063849684e-07,
+      "loss": 0.0199,
+      "step": 266
+    },
+    {
+      "epoch": 1.8287671232876712,
+      "grad_norm": 0.5267703907525366,
+      "learning_rate": 1.7977703021452185e-07,
+      "loss": 0.0282,
+      "step": 267
+    },
+    {
+      "epoch": 1.8356164383561644,
+      "grad_norm": 0.46606674643151164,
+      "learning_rate": 1.6576093197361253e-07,
+      "loss": 0.0243,
+      "step": 268
+    },
+    {
+      "epoch": 1.8424657534246576,
+      "grad_norm": 0.5978206977692369,
+      "learning_rate": 1.523044083122138e-07,
+      "loss": 0.0244,
+      "step": 269
+    },
+    {
+      "epoch": 1.8493150684931505,
+      "grad_norm": 0.6813332575839723,
+      "learning_rate": 1.39409016854693e-07,
+      "loss": 0.028,
+      "step": 270
+    },
+    {
+      "epoch": 1.856164383561644,
+      "grad_norm": 0.6027615619066997,
+      "learning_rate": 1.2707625027304104e-07,
+      "loss": 0.0252,
+      "step": 271
+    },
+    {
+      "epoch": 1.8630136986301369,
+      "grad_norm": 0.5524351842468936,
+      "learning_rate": 1.1530753611409151e-07,
+      "loss": 0.0268,
+      "step": 272
+    },
+    {
+      "epoch": 1.8698630136986303,
+      "grad_norm": 0.5640622999698867,
+      "learning_rate": 1.041042366342787e-07,
+      "loss": 0.0256,
+      "step": 273
+    },
+    {
+      "epoch": 1.8767123287671232,
+      "grad_norm": 0.536202910263522,
+      "learning_rate": 9.346764864195335e-08,
+      "loss": 0.0256,
+      "step": 274
+    },
+    {
+      "epoch": 1.8835616438356164,
+      "grad_norm": 0.5919425450976535,
+      "learning_rate": 8.339900334727536e-08,
+      "loss": 0.0315,
+      "step": 275
+    },
+    {
+      "epoch": 1.8904109589041096,
+      "grad_norm": 0.5574691459244832,
+      "learning_rate": 7.389946621969679e-08,
+      "loss": 0.0225,
+      "step": 276
+    },
+    {
+      "epoch": 1.8972602739726028,
+      "grad_norm": 0.5505059019532733,
+      "learning_rate": 6.497013685305586e-08,
+      "loss": 0.0205,
+      "step": 277
+    },
+    {
+      "epoch": 1.904109589041096,
+      "grad_norm": 0.4886781912164835,
+      "learning_rate": 5.661204883829763e-08,
+      "loss": 0.0219,
+      "step": 278
+    },
+    {
+      "epoch": 1.910958904109589,
+      "grad_norm": 0.5635253689144598,
+      "learning_rate": 4.8826169643832464e-08,
+      "loss": 0.0269,
+      "step": 279
+    },
+    {
+      "epoch": 1.9178082191780823,
+      "grad_norm": 0.5462586475286777,
+      "learning_rate": 4.1613400503550114e-08,
+      "loss": 0.0267,
+      "step": 280
+    },
+    {
+      "epoch": 1.9246575342465753,
+      "grad_norm": 0.5451325842479025,
+      "learning_rate": 3.4974576312497564e-08,
+      "loss": 0.0272,
+      "step": 281
+    },
+    {
+      "epoch": 1.9315068493150684,
+      "grad_norm": 0.42215818334666344,
+      "learning_rate": 2.8910465530240793e-08,
+      "loss": 0.021,
+      "step": 282
+    },
+    {
+      "epoch": 1.9383561643835616,
+      "grad_norm": 0.6580299707133164,
+      "learning_rate": 2.3421770091912044e-08,
+      "loss": 0.0312,
+      "step": 283
+    },
+    {
+      "epoch": 1.9452054794520548,
+      "grad_norm": 0.6439576519711626,
+      "learning_rate": 1.850912532696092e-08,
+      "loss": 0.0268,
+      "step": 284
+    },
+    {
+      "epoch": 1.952054794520548,
+      "grad_norm": 0.5683181922618227,
+      "learning_rate": 1.4173099885610997e-08,
+      "loss": 0.0244,
+      "step": 285
+    },
+    {
+      "epoch": 1.958904109589041,
+      "grad_norm": 0.5240614569580544,
+      "learning_rate": 1.041419567303914e-08,
+      "loss": 0.0258,
+      "step": 286
+    },
+    {
+      "epoch": 1.9657534246575343,
+      "grad_norm": 0.514575631763928,
+      "learning_rate": 7.2328477912769756e-09,
+      "loss": 0.0228,
+      "step": 287
+    },
+    {
+      "epoch": 1.9726027397260273,
+      "grad_norm": 0.5015672539098924,
+      "learning_rate": 4.629424488850065e-09,
+      "loss": 0.0213,
+      "step": 288
+    },
+    {
+      "epoch": 1.9794520547945207,
+      "grad_norm": 0.5696234496920316,
+      "learning_rate": 2.604227118148117e-09,
+      "loss": 0.0232,
+      "step": 289
+    },
+    {
+      "epoch": 1.9863013698630136,
+      "grad_norm": 0.5030448428911224,
+      "learning_rate": 1.1574901005456662e-09,
+      "loss": 0.0224,
+      "step": 290
+    },
+    {
+      "epoch": 1.9931506849315068,
+      "grad_norm": 0.5733211189336573,
+      "learning_rate": 2.89380899267111e-10,
+      "loss": 0.0254,
+      "step": 291
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.4940211340324302,
+      "learning_rate": 0.0,
+      "loss": 0.0212,
+      "step": 292
+    },
+    {
+      "epoch": 2.0,
+      "step": 292,
+      "total_flos": 8378343751680.0,
+      "train_loss": 0.06280981206449948,
+      "train_runtime": 833.6474,
+      "train_samples_per_second": 5.587,
+      "train_steps_per_second": 0.35
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 292,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 5000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8378343751680.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe65efe7fe2a5a9adcdaf00960ef52a6b87150f1eccbad0f148b8a50a8a9d964
+size 7288