Upload model

.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 +1,61 @@
+---
+library_name: transformers
+license: other
+base_model: Qwen2.5-7B
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: v2
+  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. -->
+
+# v2
+
+This model is a fine-tuned version of [Qwen2.5-7B](https://huggingface.co/Qwen2.5-7B) on the tower_zhen_cot_mt_gemini_distill.sft dataset.
+
+## 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: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 64
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.56.2
+- Pytorch 2.8.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.1

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 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 44242544295936.0,
+    "train_loss": 0.5097209928399425,
+    "train_runtime": 1256.4644,
+    "train_samples_per_second": 14.995,
+    "train_steps_per_second": 0.235
+}

chat_template.jinja

@@ -0,0 +1,7 @@
+{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ '<|im_start|>system
+' + system_message + '<|im_end|>
+' }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|im_start|>user
+' + content + '<|im_end|>
+<|im_start|>assistant
+' }}{% elif message['role'] == 'assistant' %}{{ content + '<|im_end|>
+' }}{% endif %}{% endfor %}

config.json

@@ -0,0 +1,59 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 131072,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.56.2",
+  "use_cache": false,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

generation_config.json

@@ -0,0 +1,9 @@
+{
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "max_new_tokens": 2048,
+  "pad_token_id": 151643,
+  "transformers_version": "4.56.2"
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ee012c61e6d4e6bd757b58bfeaa2da4a468236105083542c920661622de2401
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f548702e8598bb65364aed734f56aeef5302db942d053b3057a288df374dc8b7
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63b9faf921c6022534c4a8e1849f5d27098d5375c85f3cb851191489e3b8dde7
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e6b3f5712bf9e6156ebf6a5018e2a7e62f36c78ac8c9587f9d20482c0272c6d
+size 1089994880

model.safetensors.index.json

@@ -0,0 +1,347 @@
+{
+  "metadata": {
+    "total_parameters": 333312,
+    "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"
+  }
+}

qwen_full_sft.yaml

@@ -0,0 +1,50 @@
+### model
+model_name_or_path: Qwen2.5-7B
+flash_attn: fa2
+
+
+### method
+stage: sft
+do_train: true
+finetuning_type: full
+deepspeed: examples/deepspeed/ds_z3_config.json
+
+### dataset
+dataset: tower_zhen_cot_mt_gemini_distill.sft
+template: chatml
+cutoff_len: 12000
+# cutoff_prompt_head: true
+# max_samples: 16000
+overwrite_cache: true
+preprocessing_num_workers: 16
+dataloader_num_workers: 4
+# max_steps: 12540
+
+# default_system: "You are Qwen, a virtual human developed by the Qwen Team, Alibaba Group, capable of perceiving auditory and visual inputs, as well as generating text and speech."
+
+### output
+output_dir: saves/qwen-7b/full/sft/v2
+logging_steps: 1
+save_steps: 2500
+plot_loss: true
+overwrite_output_dir: true
+report_to: wandb  # choices: [none, wandb, tensorboard, swanlab, mlflow]
+
+### train
+per_device_train_batch_size: 4
+gradient_accumulation_steps: 2
+learning_rate: 1.0e-5
+num_train_epochs: 1.0
+lr_scheduler_type: cosine
+warmup_ratio: 0.1
+# warmup_steps: 300
+bf16: true
+ddp_timeout: 180000000
+resume_from_checkpoint: null
+# model_accepts_loss_kwargs: true
+
+## eval
+# eval_dataset: ted_yt_training_transcription.cot5.test.audio
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 400

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,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "total_flos": 44242544295936.0,
+    "train_loss": 0.5097209928399425,
+    "train_runtime": 1256.4644,
+    "train_samples_per_second": 14.995,
+    "train_steps_per_second": 0.235
+}

trainer_log.jsonl

@@ -0,0 +1,296 @@
+{"current_steps": 1, "total_steps": 295, "loss": 0.9899, "lr": 0.0, "epoch": 0.003395585738539898, "percentage": 0.34, "elapsed_time": "0:00:06", "remaining_time": "0:30:39"}
+{"current_steps": 2, "total_steps": 295, "loss": 0.9683, "lr": 3.3333333333333335e-07, "epoch": 0.006791171477079796, "percentage": 0.68, "elapsed_time": "0:00:10", "remaining_time": "0:25:09"}
+{"current_steps": 3, "total_steps": 295, "loss": 0.9722, "lr": 6.666666666666667e-07, "epoch": 0.010186757215619695, "percentage": 1.02, "elapsed_time": "0:00:13", "remaining_time": "0:22:04"}
+{"current_steps": 4, "total_steps": 295, "loss": 0.9667, "lr": 1.0000000000000002e-06, "epoch": 0.013582342954159592, "percentage": 1.36, "elapsed_time": "0:00:17", "remaining_time": "0:21:00"}
+{"current_steps": 5, "total_steps": 295, "loss": 0.9814, "lr": 1.3333333333333334e-06, "epoch": 0.01697792869269949, "percentage": 1.69, "elapsed_time": "0:00:21", "remaining_time": "0:20:23"}
+{"current_steps": 6, "total_steps": 295, "loss": 0.9502, "lr": 1.6666666666666667e-06, "epoch": 0.02037351443123939, "percentage": 2.03, "elapsed_time": "0:00:25", "remaining_time": "0:20:09"}
+{"current_steps": 7, "total_steps": 295, "loss": 0.9411, "lr": 2.0000000000000003e-06, "epoch": 0.023769100169779286, "percentage": 2.37, "elapsed_time": "0:00:30", "remaining_time": "0:20:41"}
+{"current_steps": 8, "total_steps": 295, "loss": 0.9075, "lr": 2.3333333333333336e-06, "epoch": 0.027164685908319185, "percentage": 2.71, "elapsed_time": "0:00:34", "remaining_time": "0:20:51"}
+{"current_steps": 9, "total_steps": 295, "loss": 0.8979, "lr": 2.666666666666667e-06, "epoch": 0.030560271646859084, "percentage": 3.05, "elapsed_time": "0:00:38", "remaining_time": "0:20:31"}
+{"current_steps": 10, "total_steps": 295, "loss": 0.8589, "lr": 3e-06, "epoch": 0.03395585738539898, "percentage": 3.39, "elapsed_time": "0:00:42", "remaining_time": "0:20:23"}
+{"current_steps": 11, "total_steps": 295, "loss": 0.8797, "lr": 3.3333333333333333e-06, "epoch": 0.03735144312393888, "percentage": 3.73, "elapsed_time": "0:00:46", "remaining_time": "0:20:06"}
+{"current_steps": 12, "total_steps": 295, "loss": 0.785, "lr": 3.6666666666666666e-06, "epoch": 0.04074702886247878, "percentage": 4.07, "elapsed_time": "0:00:50", "remaining_time": "0:19:53"}
+{"current_steps": 13, "total_steps": 295, "loss": 0.7932, "lr": 4.000000000000001e-06, "epoch": 0.044142614601018676, "percentage": 4.41, "elapsed_time": "0:00:54", "remaining_time": "0:19:36"}
+{"current_steps": 14, "total_steps": 295, "loss": 0.7878, "lr": 4.333333333333334e-06, "epoch": 0.04753820033955857, "percentage": 4.75, "elapsed_time": "0:00:58", "remaining_time": "0:19:27"}
+{"current_steps": 15, "total_steps": 295, "loss": 0.7854, "lr": 4.666666666666667e-06, "epoch": 0.050933786078098474, "percentage": 5.08, "elapsed_time": "0:01:01", "remaining_time": "0:19:11"}
+{"current_steps": 16, "total_steps": 295, "loss": 0.7194, "lr": 5e-06, "epoch": 0.05432937181663837, "percentage": 5.42, "elapsed_time": "0:01:07", "remaining_time": "0:19:34"}
+{"current_steps": 17, "total_steps": 295, "loss": 0.7059, "lr": 5.333333333333334e-06, "epoch": 0.057724957555178265, "percentage": 5.76, "elapsed_time": "0:01:10", "remaining_time": "0:19:19"}
+{"current_steps": 18, "total_steps": 295, "loss": 0.7236, "lr": 5.666666666666667e-06, "epoch": 0.06112054329371817, "percentage": 6.1, "elapsed_time": "0:01:14", "remaining_time": "0:19:07"}
+{"current_steps": 19, "total_steps": 295, "loss": 0.7133, "lr": 6e-06, "epoch": 0.06451612903225806, "percentage": 6.44, "elapsed_time": "0:01:20", "remaining_time": "0:19:31"}
+{"current_steps": 20, "total_steps": 295, "loss": 0.6807, "lr": 6.333333333333333e-06, "epoch": 0.06791171477079797, "percentage": 6.78, "elapsed_time": "0:01:23", "remaining_time": "0:19:14"}
+{"current_steps": 21, "total_steps": 295, "loss": 0.6717, "lr": 6.666666666666667e-06, "epoch": 0.07130730050933787, "percentage": 7.12, "elapsed_time": "0:01:28", "remaining_time": "0:19:08"}
+{"current_steps": 22, "total_steps": 295, "loss": 0.6836, "lr": 7e-06, "epoch": 0.07470288624787776, "percentage": 7.46, "elapsed_time": "0:01:31", "remaining_time": "0:18:58"}
+{"current_steps": 23, "total_steps": 295, "loss": 0.656, "lr": 7.333333333333333e-06, "epoch": 0.07809847198641766, "percentage": 7.8, "elapsed_time": "0:01:35", "remaining_time": "0:18:47"}
+{"current_steps": 24, "total_steps": 295, "loss": 0.652, "lr": 7.666666666666667e-06, "epoch": 0.08149405772495756, "percentage": 8.14, "elapsed_time": "0:01:39", "remaining_time": "0:18:39"}
+{"current_steps": 25, "total_steps": 295, "loss": 0.6237, "lr": 8.000000000000001e-06, "epoch": 0.08488964346349745, "percentage": 8.47, "elapsed_time": "0:01:42", "remaining_time": "0:18:32"}
+{"current_steps": 26, "total_steps": 295, "loss": 0.5999, "lr": 8.333333333333334e-06, "epoch": 0.08828522920203735, "percentage": 8.81, "elapsed_time": "0:01:48", "remaining_time": "0:18:47"}
+{"current_steps": 27, "total_steps": 295, "loss": 0.6229, "lr": 8.666666666666668e-06, "epoch": 0.09168081494057725, "percentage": 9.15, "elapsed_time": "0:01:52", "remaining_time": "0:18:37"}
+{"current_steps": 28, "total_steps": 295, "loss": 0.5909, "lr": 9e-06, "epoch": 0.09507640067911714, "percentage": 9.49, "elapsed_time": "0:01:56", "remaining_time": "0:18:26"}
+{"current_steps": 29, "total_steps": 295, "loss": 0.6236, "lr": 9.333333333333334e-06, "epoch": 0.09847198641765705, "percentage": 9.83, "elapsed_time": "0:01:59", "remaining_time": "0:18:17"}
+{"current_steps": 30, "total_steps": 295, "loss": 0.5978, "lr": 9.666666666666667e-06, "epoch": 0.10186757215619695, "percentage": 10.17, "elapsed_time": "0:02:03", "remaining_time": "0:18:10"}
+{"current_steps": 31, "total_steps": 295, "loss": 0.6206, "lr": 1e-05, "epoch": 0.10526315789473684, "percentage": 10.51, "elapsed_time": "0:02:06", "remaining_time": "0:18:00"}
+{"current_steps": 32, "total_steps": 295, "loss": 0.6103, "lr": 9.999648647603774e-06, "epoch": 0.10865874363327674, "percentage": 10.85, "elapsed_time": "0:02:10", "remaining_time": "0:17:51"}
+{"current_steps": 33, "total_steps": 295, "loss": 0.5823, "lr": 9.998594639794502e-06, "epoch": 0.11205432937181664, "percentage": 11.19, "elapsed_time": "0:02:14", "remaining_time": "0:17:45"}
+{"current_steps": 34, "total_steps": 295, "loss": 0.5724, "lr": 9.996838124703448e-06, "epoch": 0.11544991511035653, "percentage": 11.53, "elapsed_time": "0:02:19", "remaining_time": "0:17:47"}
+{"current_steps": 35, "total_steps": 295, "loss": 0.6006, "lr": 9.994379349192927e-06, "epoch": 0.11884550084889643, "percentage": 11.86, "elapsed_time": "0:02:22", "remaining_time": "0:17:39"}
+{"current_steps": 36, "total_steps": 295, "loss": 0.5834, "lr": 9.991218658821609e-06, "epoch": 0.12224108658743633, "percentage": 12.2, "elapsed_time": "0:02:26", "remaining_time": "0:17:31"}
+{"current_steps": 37, "total_steps": 295, "loss": 0.5781, "lr": 9.987356497795944e-06, "epoch": 0.12563667232597622, "percentage": 12.54, "elapsed_time": "0:02:31", "remaining_time": "0:17:36"}
+{"current_steps": 38, "total_steps": 295, "loss": 0.5613, "lr": 9.982793408907747e-06, "epoch": 0.12903225806451613, "percentage": 12.88, "elapsed_time": "0:02:35", "remaining_time": "0:17:29"}
+{"current_steps": 39, "total_steps": 295, "loss": 0.5518, "lr": 9.977530033457906e-06, "epoch": 0.13242784380305603, "percentage": 13.22, "elapsed_time": "0:02:39", "remaining_time": "0:17:24"}
+{"current_steps": 40, "total_steps": 295, "loss": 0.5875, "lr": 9.971567111166246e-06, "epoch": 0.13582342954159593, "percentage": 13.56, "elapsed_time": "0:02:42", "remaining_time": "0:17:16"}
+{"current_steps": 41, "total_steps": 295, "loss": 0.5666, "lr": 9.964905480067585e-06, "epoch": 0.13921901528013583, "percentage": 13.9, "elapsed_time": "0:02:46", "remaining_time": "0:17:11"}
+{"current_steps": 42, "total_steps": 295, "loss": 0.5829, "lr": 9.957546076393944e-06, "epoch": 0.14261460101867574, "percentage": 14.24, "elapsed_time": "0:02:51", "remaining_time": "0:17:15"}
+{"current_steps": 43, "total_steps": 295, "loss": 0.592, "lr": 9.949489934442966e-06, "epoch": 0.1460101867572156, "percentage": 14.58, "elapsed_time": "0:02:55", "remaining_time": "0:17:08"}
+{"current_steps": 44, "total_steps": 295, "loss": 0.554, "lr": 9.940738186432565e-06, "epoch": 0.1494057724957555, "percentage": 14.92, "elapsed_time": "0:02:58", "remaining_time": "0:17:00"}
+{"current_steps": 45, "total_steps": 295, "loss": 0.5563, "lr": 9.931292062341793e-06, "epoch": 0.15280135823429541, "percentage": 15.25, "elapsed_time": "0:03:02", "remaining_time": "0:16:54"}
+{"current_steps": 46, "total_steps": 295, "loss": 0.5349, "lr": 9.921152889737985e-06, "epoch": 0.15619694397283532, "percentage": 15.59, "elapsed_time": "0:03:07", "remaining_time": "0:16:52"}
+{"current_steps": 47, "total_steps": 295, "loss": 0.5885, "lr": 9.910322093590177e-06, "epoch": 0.15959252971137522, "percentage": 15.93, "elapsed_time": "0:03:12", "remaining_time": "0:16:54"}
+{"current_steps": 48, "total_steps": 295, "loss": 0.555, "lr": 9.898801196068839e-06, "epoch": 0.16298811544991512, "percentage": 16.27, "elapsed_time": "0:03:19", "remaining_time": "0:17:04"}
+{"current_steps": 49, "total_steps": 295, "loss": 0.5412, "lr": 9.886591816331953e-06, "epoch": 0.166383701188455, "percentage": 16.61, "elapsed_time": "0:03:23", "remaining_time": "0:17:01"}
+{"current_steps": 50, "total_steps": 295, "loss": 0.5266, "lr": 9.87369567029745e-06, "epoch": 0.1697792869269949, "percentage": 16.95, "elapsed_time": "0:03:28", "remaining_time": "0:17:00"}
+{"current_steps": 51, "total_steps": 295, "loss": 0.5364, "lr": 9.860114570402055e-06, "epoch": 0.1731748726655348, "percentage": 17.29, "elapsed_time": "0:03:31", "remaining_time": "0:16:53"}
+{"current_steps": 52, "total_steps": 295, "loss": 0.544, "lr": 9.845850425346563e-06, "epoch": 0.1765704584040747, "percentage": 17.63, "elapsed_time": "0:03:37", "remaining_time": "0:16:54"}
+{"current_steps": 53, "total_steps": 295, "loss": 0.526, "lr": 9.830905239827592e-06, "epoch": 0.1799660441426146, "percentage": 17.97, "elapsed_time": "0:03:42", "remaining_time": "0:16:54"}
+{"current_steps": 54, "total_steps": 295, "loss": 0.5384, "lr": 9.815281114255841e-06, "epoch": 0.1833616298811545, "percentage": 18.31, "elapsed_time": "0:03:48", "remaining_time": "0:17:00"}
+{"current_steps": 55, "total_steps": 295, "loss": 0.5216, "lr": 9.798980244460892e-06, "epoch": 0.1867572156196944, "percentage": 18.64, "elapsed_time": "0:03:52", "remaining_time": "0:16:53"}
+{"current_steps": 56, "total_steps": 295, "loss": 0.5602, "lr": 9.782004921382612e-06, "epoch": 0.19015280135823429, "percentage": 18.98, "elapsed_time": "0:03:56", "remaining_time": "0:16:48"}
+{"current_steps": 57, "total_steps": 295, "loss": 0.5211, "lr": 9.764357530749178e-06, "epoch": 0.1935483870967742, "percentage": 19.32, "elapsed_time": "0:03:59", "remaining_time": "0:16:41"}
+{"current_steps": 58, "total_steps": 295, "loss": 0.5408, "lr": 9.74604055274178e-06, "epoch": 0.1969439728353141, "percentage": 19.66, "elapsed_time": "0:04:03", "remaining_time": "0:16:35"}
+{"current_steps": 59, "total_steps": 295, "loss": 0.5297, "lr": 9.727056561646067e-06, "epoch": 0.200339558573854, "percentage": 20.0, "elapsed_time": "0:04:06", "remaining_time": "0:16:27"}
+{"current_steps": 60, "total_steps": 295, "loss": 0.5241, "lr": 9.707408225490343e-06, "epoch": 0.2037351443123939, "percentage": 20.34, "elapsed_time": "0:04:10", "remaining_time": "0:16:22"}
+{"current_steps": 61, "total_steps": 295, "loss": 0.5373, "lr": 9.687098305670606e-06, "epoch": 0.2071307300509338, "percentage": 20.68, "elapsed_time": "0:04:15", "remaining_time": "0:16:18"}
+{"current_steps": 62, "total_steps": 295, "loss": 0.5397, "lr": 9.66612965656245e-06, "epoch": 0.21052631578947367, "percentage": 21.02, "elapsed_time": "0:04:18", "remaining_time": "0:16:12"}
+{"current_steps": 63, "total_steps": 295, "loss": 0.5444, "lr": 9.644505225119922e-06, "epoch": 0.21392190152801357, "percentage": 21.36, "elapsed_time": "0:04:22", "remaining_time": "0:16:05"}
+{"current_steps": 64, "total_steps": 295, "loss": 0.4984, "lr": 9.622228050461345e-06, "epoch": 0.21731748726655348, "percentage": 21.69, "elapsed_time": "0:04:27", "remaining_time": "0:16:03"}
+{"current_steps": 65, "total_steps": 295, "loss": 0.5214, "lr": 9.599301263442194e-06, "epoch": 0.22071307300509338, "percentage": 22.03, "elapsed_time": "0:04:30", "remaining_time": "0:15:58"}
+{"current_steps": 66, "total_steps": 295, "loss": 0.5265, "lr": 9.575728086215093e-06, "epoch": 0.22410865874363328, "percentage": 22.37, "elapsed_time": "0:04:34", "remaining_time": "0:15:52"}
+{"current_steps": 67, "total_steps": 295, "loss": 0.5155, "lr": 9.551511831776966e-06, "epoch": 0.22750424448217318, "percentage": 22.71, "elapsed_time": "0:04:38", "remaining_time": "0:15:46"}
+{"current_steps": 68, "total_steps": 295, "loss": 0.5277, "lr": 9.526655903503423e-06, "epoch": 0.23089983022071306, "percentage": 23.05, "elapsed_time": "0:04:42", "remaining_time": "0:15:41"}
+{"current_steps": 69, "total_steps": 295, "loss": 0.5307, "lr": 9.501163794670445e-06, "epoch": 0.23429541595925296, "percentage": 23.39, "elapsed_time": "0:04:47", "remaining_time": "0:15:40"}
+{"current_steps": 70, "total_steps": 295, "loss": 0.51, "lr": 9.475039087963443e-06, "epoch": 0.23769100169779286, "percentage": 23.73, "elapsed_time": "0:04:50", "remaining_time": "0:15:34"}
+{"current_steps": 71, "total_steps": 295, "loss": 0.517, "lr": 9.448285454973739e-06, "epoch": 0.24108658743633277, "percentage": 24.07, "elapsed_time": "0:04:54", "remaining_time": "0:15:28"}
+{"current_steps": 72, "total_steps": 295, "loss": 0.5223, "lr": 9.420906655682553e-06, "epoch": 0.24448217317487267, "percentage": 24.41, "elapsed_time": "0:04:57", "remaining_time": "0:15:22"}
+{"current_steps": 73, "total_steps": 295, "loss": 0.5257, "lr": 9.392906537932582e-06, "epoch": 0.24787775891341257, "percentage": 24.75, "elapsed_time": "0:05:01", "remaining_time": "0:15:16"}
+{"current_steps": 74, "total_steps": 295, "loss": 0.5187, "lr": 9.364289036887214e-06, "epoch": 0.25127334465195245, "percentage": 25.08, "elapsed_time": "0:05:05", "remaining_time": "0:15:11"}
+{"current_steps": 75, "total_steps": 295, "loss": 0.5127, "lr": 9.335058174477472e-06, "epoch": 0.2546689303904924, "percentage": 25.42, "elapsed_time": "0:05:08", "remaining_time": "0:15:05"}
+{"current_steps": 76, "total_steps": 295, "loss": 0.4966, "lr": 9.305218058836778e-06, "epoch": 0.25806451612903225, "percentage": 25.76, "elapsed_time": "0:05:14", "remaining_time": "0:15:05"}
+{"current_steps": 77, "total_steps": 295, "loss": 0.5265, "lr": 9.274772883723587e-06, "epoch": 0.2614601018675722, "percentage": 26.1, "elapsed_time": "0:05:17", "remaining_time": "0:14:59"}
+{"current_steps": 78, "total_steps": 295, "loss": 0.5003, "lr": 9.24372692793199e-06, "epoch": 0.26485568760611206, "percentage": 26.44, "elapsed_time": "0:05:22", "remaining_time": "0:14:56"}
+{"current_steps": 79, "total_steps": 295, "loss": 0.4971, "lr": 9.21208455469037e-06, "epoch": 0.26825127334465193, "percentage": 26.78, "elapsed_time": "0:05:25", "remaining_time": "0:14:50"}
+{"current_steps": 80, "total_steps": 295, "loss": 0.491, "lr": 9.179850211048193e-06, "epoch": 0.27164685908319186, "percentage": 27.12, "elapsed_time": "0:05:29", "remaining_time": "0:14:46"}
+{"current_steps": 81, "total_steps": 295, "loss": 0.5033, "lr": 9.14702842725101e-06, "epoch": 0.27504244482173174, "percentage": 27.46, "elapsed_time": "0:05:33", "remaining_time": "0:14:41"}
+{"current_steps": 82, "total_steps": 295, "loss": 0.5157, "lr": 9.113623816103775e-06, "epoch": 0.27843803056027167, "percentage": 27.8, "elapsed_time": "0:05:37", "remaining_time": "0:14:36"}
+{"current_steps": 83, "total_steps": 295, "loss": 0.5103, "lr": 9.079641072322555e-06, "epoch": 0.28183361629881154, "percentage": 28.14, "elapsed_time": "0:05:41", "remaining_time": "0:14:31"}
+{"current_steps": 84, "total_steps": 295, "loss": 0.5118, "lr": 9.045084971874738e-06, "epoch": 0.28522920203735147, "percentage": 28.47, "elapsed_time": "0:05:44", "remaining_time": "0:14:26"}
+{"current_steps": 85, "total_steps": 295, "loss": 0.4753, "lr": 9.009960371307798e-06, "epoch": 0.28862478777589134, "percentage": 28.81, "elapsed_time": "0:05:49", "remaining_time": "0:14:22"}
+{"current_steps": 86, "total_steps": 295, "loss": 0.5065, "lr": 8.974272207066767e-06, "epoch": 0.2920203735144312, "percentage": 29.15, "elapsed_time": "0:05:52", "remaining_time": "0:14:17"}
+{"current_steps": 87, "total_steps": 295, "loss": 0.496, "lr": 8.938025494800454e-06, "epoch": 0.29541595925297115, "percentage": 29.49, "elapsed_time": "0:05:56", "remaining_time": "0:14:12"}
+{"current_steps": 88, "total_steps": 295, "loss": 0.4768, "lr": 8.901225328656543e-06, "epoch": 0.298811544991511, "percentage": 29.83, "elapsed_time": "0:06:02", "remaining_time": "0:14:12"}
+{"current_steps": 89, "total_steps": 295, "loss": 0.4982, "lr": 8.863876880565656e-06, "epoch": 0.30220713073005095, "percentage": 30.17, "elapsed_time": "0:06:06", "remaining_time": "0:14:07"}
+{"current_steps": 90, "total_steps": 295, "loss": 0.49, "lr": 8.825985399514488e-06, "epoch": 0.30560271646859083, "percentage": 30.51, "elapsed_time": "0:06:10", "remaining_time": "0:14:04"}
+{"current_steps": 91, "total_steps": 295, "loss": 0.508, "lr": 8.787556210808101e-06, "epoch": 0.3089983022071307, "percentage": 30.85, "elapsed_time": "0:06:14", "remaining_time": "0:13:58"}
+{"current_steps": 92, "total_steps": 295, "loss": 0.493, "lr": 8.748594715321512e-06, "epoch": 0.31239388794567063, "percentage": 31.19, "elapsed_time": "0:06:20", "remaining_time": "0:13:58"}
+{"current_steps": 93, "total_steps": 295, "loss": 0.5063, "lr": 8.70910638874064e-06, "epoch": 0.3157894736842105, "percentage": 31.53, "elapsed_time": "0:06:23", "remaining_time": "0:13:52"}
+{"current_steps": 94, "total_steps": 295, "loss": 0.4953, "lr": 8.669096780792754e-06, "epoch": 0.31918505942275044, "percentage": 31.86, "elapsed_time": "0:06:26", "remaining_time": "0:13:47"}
+{"current_steps": 95, "total_steps": 295, "loss": 0.5062, "lr": 8.628571514466502e-06, "epoch": 0.3225806451612903, "percentage": 32.2, "elapsed_time": "0:06:30", "remaining_time": "0:13:41"}
+{"current_steps": 96, "total_steps": 295, "loss": 0.5254, "lr": 8.587536285221656e-06, "epoch": 0.32597623089983024, "percentage": 32.54, "elapsed_time": "0:06:34", "remaining_time": "0:13:38"}
+{"current_steps": 97, "total_steps": 295, "loss": 0.5107, "lr": 8.545996860188668e-06, "epoch": 0.3293718166383701, "percentage": 32.88, "elapsed_time": "0:06:38", "remaining_time": "0:13:33"}
+{"current_steps": 98, "total_steps": 295, "loss": 0.4767, "lr": 8.503959077358143e-06, "epoch": 0.33276740237691, "percentage": 33.22, "elapsed_time": "0:06:43", "remaining_time": "0:13:30"}
+{"current_steps": 99, "total_steps": 295, "loss": 0.4991, "lr": 8.46142884476038e-06, "epoch": 0.3361629881154499, "percentage": 33.56, "elapsed_time": "0:06:46", "remaining_time": "0:13:25"}
+{"current_steps": 100, "total_steps": 295, "loss": 0.5202, "lr": 8.418412139635026e-06, "epoch": 0.3395585738539898, "percentage": 33.9, "elapsed_time": "0:06:50", "remaining_time": "0:13:20"}
+{"current_steps": 101, "total_steps": 295, "loss": 0.5019, "lr": 8.374915007591053e-06, "epoch": 0.34295415959252973, "percentage": 34.24, "elapsed_time": "0:06:53", "remaining_time": "0:13:14"}
+{"current_steps": 102, "total_steps": 295, "loss": 0.4981, "lr": 8.330943561757092e-06, "epoch": 0.3463497453310696, "percentage": 34.58, "elapsed_time": "0:06:57", "remaining_time": "0:13:09"}
+{"current_steps": 103, "total_steps": 295, "loss": 0.4828, "lr": 8.286503981922284e-06, "epoch": 0.34974533106960953, "percentage": 34.92, "elapsed_time": "0:07:02", "remaining_time": "0:13:08"}
+{"current_steps": 104, "total_steps": 295, "loss": 0.4677, "lr": 8.241602513667775e-06, "epoch": 0.3531409168081494, "percentage": 35.25, "elapsed_time": "0:07:06", "remaining_time": "0:13:04"}
+{"current_steps": 105, "total_steps": 295, "loss": 0.4991, "lr": 8.19624546748895e-06, "epoch": 0.3565365025466893, "percentage": 35.59, "elapsed_time": "0:07:10", "remaining_time": "0:12:59"}
+{"current_steps": 106, "total_steps": 295, "loss": 0.4874, "lr": 8.150439217908557e-06, "epoch": 0.3599320882852292, "percentage": 35.93, "elapsed_time": "0:07:14", "remaining_time": "0:12:54"}
+{"current_steps": 107, "total_steps": 295, "loss": 0.4613, "lr": 8.104190202580811e-06, "epoch": 0.3633276740237691, "percentage": 36.27, "elapsed_time": "0:07:21", "remaining_time": "0:12:54"}
+{"current_steps": 108, "total_steps": 295, "loss": 0.4981, "lr": 8.057504921386661e-06, "epoch": 0.366723259762309, "percentage": 36.61, "elapsed_time": "0:07:27", "remaining_time": "0:12:54"}
+{"current_steps": 109, "total_steps": 295, "loss": 0.4812, "lr": 8.010389935520269e-06, "epoch": 0.3701188455008489, "percentage": 36.95, "elapsed_time": "0:07:31", "remaining_time": "0:12:49"}
+{"current_steps": 110, "total_steps": 295, "loss": 0.4901, "lr": 7.962851866566912e-06, "epoch": 0.3735144312393888, "percentage": 37.29, "elapsed_time": "0:07:34", "remaining_time": "0:12:44"}
+{"current_steps": 111, "total_steps": 295, "loss": 0.4841, "lr": 7.914897395572362e-06, "epoch": 0.3769100169779287, "percentage": 37.63, "elapsed_time": "0:07:38", "remaining_time": "0:12:39"}
+{"current_steps": 112, "total_steps": 295, "loss": 0.4813, "lr": 7.866533262103937e-06, "epoch": 0.38030560271646857, "percentage": 37.97, "elapsed_time": "0:07:42", "remaining_time": "0:12:35"}
+{"current_steps": 113, "total_steps": 295, "loss": 0.4731, "lr": 7.817766263303312e-06, "epoch": 0.3837011884550085, "percentage": 38.31, "elapsed_time": "0:07:45", "remaining_time": "0:12:29"}
+{"current_steps": 114, "total_steps": 295, "loss": 0.4988, "lr": 7.768603252931243e-06, "epoch": 0.3870967741935484, "percentage": 38.64, "elapsed_time": "0:07:49", "remaining_time": "0:12:25"}
+{"current_steps": 115, "total_steps": 295, "loss": 0.4825, "lr": 7.719051140404327e-06, "epoch": 0.3904923599320883, "percentage": 38.98, "elapsed_time": "0:07:53", "remaining_time": "0:12:20"}
+{"current_steps": 116, "total_steps": 295, "loss": 0.4742, "lr": 7.669116889823955e-06, "epoch": 0.3938879456706282, "percentage": 39.32, "elapsed_time": "0:07:57", "remaining_time": "0:12:16"}
+{"current_steps": 117, "total_steps": 295, "loss": 0.4743, "lr": 7.6188075189975644e-06, "epoch": 0.39728353140916806, "percentage": 39.66, "elapsed_time": "0:08:00", "remaining_time": "0:12:11"}
+{"current_steps": 118, "total_steps": 295, "loss": 0.4739, "lr": 7.568130098452352e-06, "epoch": 0.400679117147708, "percentage": 40.0, "elapsed_time": "0:08:04", "remaining_time": "0:12:07"}
+{"current_steps": 119, "total_steps": 295, "loss": 0.4814, "lr": 7.517091750441576e-06, "epoch": 0.40407470288624786, "percentage": 40.34, "elapsed_time": "0:08:08", "remaining_time": "0:12:02"}
+{"current_steps": 120, "total_steps": 295, "loss": 0.4805, "lr": 7.465699647943586e-06, "epoch": 0.4074702886247878, "percentage": 40.68, "elapsed_time": "0:08:11", "remaining_time": "0:11:57"}
+{"current_steps": 121, "total_steps": 295, "loss": 0.4953, "lr": 7.413961013653725e-06, "epoch": 0.41086587436332767, "percentage": 41.02, "elapsed_time": "0:08:15", "remaining_time": "0:11:52"}
+{"current_steps": 122, "total_steps": 295, "loss": 0.4588, "lr": 7.361883118969248e-06, "epoch": 0.4142614601018676, "percentage": 41.36, "elapsed_time": "0:08:19", "remaining_time": "0:11:47"}
+{"current_steps": 123, "total_steps": 295, "loss": 0.469, "lr": 7.309473282967387e-06, "epoch": 0.41765704584040747, "percentage": 41.69, "elapsed_time": "0:08:24", "remaining_time": "0:11:45"}
+{"current_steps": 124, "total_steps": 295, "loss": 0.4812, "lr": 7.256738871376733e-06, "epoch": 0.42105263157894735, "percentage": 42.03, "elapsed_time": "0:08:27", "remaining_time": "0:11:40"}
+{"current_steps": 125, "total_steps": 295, "loss": 0.456, "lr": 7.203687295542032e-06, "epoch": 0.4244482173174873, "percentage": 42.37, "elapsed_time": "0:08:31", "remaining_time": "0:11:35"}
+{"current_steps": 126, "total_steps": 295, "loss": 0.4597, "lr": 7.1503260113826035e-06, "epoch": 0.42784380305602715, "percentage": 42.71, "elapsed_time": "0:08:34", "remaining_time": "0:11:30"}
+{"current_steps": 127, "total_steps": 295, "loss": 0.4884, "lr": 7.09666251834447e-06, "epoch": 0.4312393887945671, "percentage": 43.05, "elapsed_time": "0:08:38", "remaining_time": "0:11:25"}
+{"current_steps": 128, "total_steps": 295, "loss": 0.4581, "lr": 7.042704358346375e-06, "epoch": 0.43463497453310695, "percentage": 43.39, "elapsed_time": "0:08:42", "remaining_time": "0:11:21"}
+{"current_steps": 129, "total_steps": 295, "loss": 0.4672, "lr": 6.988459114719849e-06, "epoch": 0.4380305602716469, "percentage": 43.73, "elapsed_time": "0:08:46", "remaining_time": "0:11:17"}
+{"current_steps": 130, "total_steps": 295, "loss": 0.4524, "lr": 6.933934411143419e-06, "epoch": 0.44142614601018676, "percentage": 44.07, "elapsed_time": "0:08:50", "remaining_time": "0:11:13"}
+{"current_steps": 131, "total_steps": 295, "loss": 0.4738, "lr": 6.879137910571191e-06, "epoch": 0.44482173174872663, "percentage": 44.41, "elapsed_time": "0:08:54", "remaining_time": "0:11:08"}
+{"current_steps": 132, "total_steps": 295, "loss": 0.4679, "lr": 6.824077314155877e-06, "epoch": 0.44821731748726656, "percentage": 44.75, "elapsed_time": "0:08:59", "remaining_time": "0:11:05"}
+{"current_steps": 133, "total_steps": 295, "loss": 0.4909, "lr": 6.768760360166471e-06, "epoch": 0.45161290322580644, "percentage": 45.08, "elapsed_time": "0:09:02", "remaining_time": "0:11:00"}
+{"current_steps": 134, "total_steps": 295, "loss": 0.4734, "lr": 6.713194822900707e-06, "epoch": 0.45500848896434637, "percentage": 45.42, "elapsed_time": "0:09:06", "remaining_time": "0:10:56"}
+{"current_steps": 135, "total_steps": 295, "loss": 0.4979, "lr": 6.657388511592453e-06, "epoch": 0.45840407470288624, "percentage": 45.76, "elapsed_time": "0:09:11", "remaining_time": "0:10:53"}
+{"current_steps": 136, "total_steps": 295, "loss": 0.4826, "lr": 6.601349269314188e-06, "epoch": 0.4617996604414261, "percentage": 46.1, "elapsed_time": "0:09:14", "remaining_time": "0:10:48"}
+{"current_steps": 137, "total_steps": 295, "loss": 0.4685, "lr": 6.545084971874738e-06, "epoch": 0.46519524617996605, "percentage": 46.44, "elapsed_time": "0:09:18", "remaining_time": "0:10:44"}
+{"current_steps": 138, "total_steps": 295, "loss": 0.483, "lr": 6.488603526712391e-06, "epoch": 0.4685908319185059, "percentage": 46.78, "elapsed_time": "0:09:22", "remaining_time": "0:10:39"}
+{"current_steps": 139, "total_steps": 295, "loss": 0.5028, "lr": 6.431912871783587e-06, "epoch": 0.47198641765704585, "percentage": 47.12, "elapsed_time": "0:09:25", "remaining_time": "0:10:34"}
+{"current_steps": 140, "total_steps": 295, "loss": 0.4669, "lr": 6.3750209744473105e-06, "epoch": 0.47538200339558573, "percentage": 47.46, "elapsed_time": "0:09:29", "remaining_time": "0:10:30"}
+{"current_steps": 141, "total_steps": 295, "loss": 0.4735, "lr": 6.3179358303453386e-06, "epoch": 0.47877758913412566, "percentage": 47.8, "elapsed_time": "0:09:34", "remaining_time": "0:10:27"}
+{"current_steps": 142, "total_steps": 295, "loss": 0.5051, "lr": 6.260665462278544e-06, "epoch": 0.48217317487266553, "percentage": 48.14, "elapsed_time": "0:09:37", "remaining_time": "0:10:22"}
+{"current_steps": 143, "total_steps": 295, "loss": 0.4762, "lr": 6.203217919079343e-06, "epoch": 0.4855687606112054, "percentage": 48.47, "elapsed_time": "0:09:41", "remaining_time": "0:10:18"}
+{"current_steps": 144, "total_steps": 295, "loss": 0.4869, "lr": 6.145601274480521e-06, "epoch": 0.48896434634974534, "percentage": 48.81, "elapsed_time": "0:09:44", "remaining_time": "0:10:13"}
+{"current_steps": 145, "total_steps": 295, "loss": 0.4611, "lr": 6.08782362598054e-06, "epoch": 0.4923599320882852, "percentage": 49.15, "elapsed_time": "0:09:49", "remaining_time": "0:10:09"}
+{"current_steps": 146, "total_steps": 295, "loss": 0.4763, "lr": 6.029893093705492e-06, "epoch": 0.49575551782682514, "percentage": 49.49, "elapsed_time": "0:09:52", "remaining_time": "0:10:05"}
+{"current_steps": 147, "total_steps": 295, "loss": 0.4654, "lr": 5.971817819267914e-06, "epoch": 0.499151103565365, "percentage": 49.83, "elapsed_time": "0:09:56", "remaining_time": "0:10:00"}
+{"current_steps": 148, "total_steps": 295, "loss": 0.5044, "lr": 5.9136059646225375e-06, "epoch": 0.5025466893039049, "percentage": 50.17, "elapsed_time": "0:10:00", "remaining_time": "0:09:56"}
+{"current_steps": 149, "total_steps": 295, "loss": 0.4435, "lr": 5.855265710919211e-06, "epoch": 0.5059422750424448, "percentage": 50.51, "elapsed_time": "0:10:05", "remaining_time": "0:09:53"}
+{"current_steps": 150, "total_steps": 295, "loss": 0.4834, "lr": 5.796805257353109e-06, "epoch": 0.5093378607809848, "percentage": 50.85, "elapsed_time": "0:10:13", "remaining_time": "0:09:52"}
+{"current_steps": 151, "total_steps": 295, "loss": 0.461, "lr": 5.738232820012407e-06, "epoch": 0.5127334465195246, "percentage": 51.19, "elapsed_time": "0:10:16", "remaining_time": "0:09:48"}
+{"current_steps": 152, "total_steps": 295, "loss": 0.4611, "lr": 5.679556630723592e-06, "epoch": 0.5161290322580645, "percentage": 51.53, "elapsed_time": "0:10:20", "remaining_time": "0:09:43"}
+{"current_steps": 153, "total_steps": 295, "loss": 0.4712, "lr": 5.620784935894548e-06, "epoch": 0.5195246179966044, "percentage": 51.86, "elapsed_time": "0:10:23", "remaining_time": "0:09:38"}
+{"current_steps": 154, "total_steps": 295, "loss": 0.4906, "lr": 5.561925995355595e-06, "epoch": 0.5229202037351444, "percentage": 52.2, "elapsed_time": "0:10:29", "remaining_time": "0:09:36"}
+{"current_steps": 155, "total_steps": 295, "loss": 0.4836, "lr": 5.5029880811986546e-06, "epoch": 0.5263157894736842, "percentage": 52.54, "elapsed_time": "0:10:33", "remaining_time": "0:09:32"}
+{"current_steps": 156, "total_steps": 295, "loss": 0.4568, "lr": 5.443979476614674e-06, "epoch": 0.5297113752122241, "percentage": 52.88, "elapsed_time": "0:10:37", "remaining_time": "0:09:27"}
+{"current_steps": 157, "total_steps": 295, "loss": 0.4669, "lr": 5.384908474729501e-06, "epoch": 0.533106960950764, "percentage": 53.22, "elapsed_time": "0:10:40", "remaining_time": "0:09:23"}
+{"current_steps": 158, "total_steps": 295, "loss": 0.4631, "lr": 5.325783377438357e-06, "epoch": 0.5365025466893039, "percentage": 53.56, "elapsed_time": "0:10:45", "remaining_time": "0:09:19"}
+{"current_steps": 159, "total_steps": 295, "loss": 0.4824, "lr": 5.266612494239088e-06, "epoch": 0.5398981324278438, "percentage": 53.9, "elapsed_time": "0:10:50", "remaining_time": "0:09:16"}
+{"current_steps": 160, "total_steps": 295, "loss": 0.4619, "lr": 5.207404141064334e-06, "epoch": 0.5432937181663837, "percentage": 54.24, "elapsed_time": "0:10:56", "remaining_time": "0:09:13"}
+{"current_steps": 161, "total_steps": 295, "loss": 0.518, "lr": 5.148166639112799e-06, "epoch": 0.5466893039049237, "percentage": 54.58, "elapsed_time": "0:11:00", "remaining_time": "0:09:10"}
+{"current_steps": 162, "total_steps": 295, "loss": 0.4567, "lr": 5.088908313679788e-06, "epoch": 0.5500848896434635, "percentage": 54.92, "elapsed_time": "0:11:04", "remaining_time": "0:09:05"}
+{"current_steps": 163, "total_steps": 295, "loss": 0.4649, "lr": 5.029637492987153e-06, "epoch": 0.5534804753820034, "percentage": 55.25, "elapsed_time": "0:11:08", "remaining_time": "0:09:01"}
+{"current_steps": 164, "total_steps": 295, "loss": 0.4682, "lr": 4.970362507012848e-06, "epoch": 0.5568760611205433, "percentage": 55.59, "elapsed_time": "0:11:12", "remaining_time": "0:08:57"}
+{"current_steps": 165, "total_steps": 295, "loss": 0.4594, "lr": 4.911091686320213e-06, "epoch": 0.5602716468590832, "percentage": 55.93, "elapsed_time": "0:11:16", "remaining_time": "0:08:52"}
+{"current_steps": 166, "total_steps": 295, "loss": 0.4266, "lr": 4.8518333608872015e-06, "epoch": 0.5636672325976231, "percentage": 56.27, "elapsed_time": "0:11:19", "remaining_time": "0:08:48"}
+{"current_steps": 167, "total_steps": 295, "loss": 0.4745, "lr": 4.792595858935668e-06, "epoch": 0.567062818336163, "percentage": 56.61, "elapsed_time": "0:11:23", "remaining_time": "0:08:43"}
+{"current_steps": 168, "total_steps": 295, "loss": 0.4863, "lr": 4.733387505760913e-06, "epoch": 0.5704584040747029, "percentage": 56.95, "elapsed_time": "0:11:27", "remaining_time": "0:08:39"}
+{"current_steps": 169, "total_steps": 295, "loss": 0.4459, "lr": 4.674216622561645e-06, "epoch": 0.5738539898132428, "percentage": 57.29, "elapsed_time": "0:11:31", "remaining_time": "0:08:35"}
+{"current_steps": 170, "total_steps": 295, "loss": 0.4571, "lr": 4.6150915252705005e-06, "epoch": 0.5772495755517827, "percentage": 57.63, "elapsed_time": "0:11:35", "remaining_time": "0:08:31"}
+{"current_steps": 171, "total_steps": 295, "loss": 0.4701, "lr": 4.556020523385326e-06, "epoch": 0.5806451612903226, "percentage": 57.97, "elapsed_time": "0:11:38", "remaining_time": "0:08:26"}
+{"current_steps": 172, "total_steps": 295, "loss": 0.477, "lr": 4.497011918801347e-06, "epoch": 0.5840407470288624, "percentage": 58.31, "elapsed_time": "0:11:43", "remaining_time": "0:08:23"}
+{"current_steps": 173, "total_steps": 295, "loss": 0.4963, "lr": 4.438074004644407e-06, "epoch": 0.5874363327674024, "percentage": 58.64, "elapsed_time": "0:11:47", "remaining_time": "0:08:18"}
+{"current_steps": 174, "total_steps": 295, "loss": 0.4502, "lr": 4.379215064105454e-06, "epoch": 0.5908319185059423, "percentage": 58.98, "elapsed_time": "0:11:50", "remaining_time": "0:08:14"}
+{"current_steps": 175, "total_steps": 295, "loss": 0.4667, "lr": 4.32044336927641e-06, "epoch": 0.5942275042444821, "percentage": 59.32, "elapsed_time": "0:11:55", "remaining_time": "0:08:10"}
+{"current_steps": 176, "total_steps": 295, "loss": 0.469, "lr": 4.261767179987595e-06, "epoch": 0.597623089983022, "percentage": 59.66, "elapsed_time": "0:11:59", "remaining_time": "0:08:06"}
+{"current_steps": 177, "total_steps": 295, "loss": 0.4549, "lr": 4.203194742646893e-06, "epoch": 0.601018675721562, "percentage": 60.0, "elapsed_time": "0:12:03", "remaining_time": "0:08:02"}
+{"current_steps": 178, "total_steps": 295, "loss": 0.4589, "lr": 4.1447342890807905e-06, "epoch": 0.6044142614601019, "percentage": 60.34, "elapsed_time": "0:12:07", "remaining_time": "0:07:57"}
+{"current_steps": 179, "total_steps": 295, "loss": 0.4525, "lr": 4.086394035377463e-06, "epoch": 0.6078098471986417, "percentage": 60.68, "elapsed_time": "0:12:10", "remaining_time": "0:07:53"}
+{"current_steps": 180, "total_steps": 295, "loss": 0.4652, "lr": 4.028182180732088e-06, "epoch": 0.6112054329371817, "percentage": 61.02, "elapsed_time": "0:12:17", "remaining_time": "0:07:51"}
+{"current_steps": 181, "total_steps": 295, "loss": 0.4887, "lr": 3.970106906294509e-06, "epoch": 0.6146010186757216, "percentage": 61.36, "elapsed_time": "0:12:22", "remaining_time": "0:07:47"}
+{"current_steps": 182, "total_steps": 295, "loss": 0.4928, "lr": 3.912176374019462e-06, "epoch": 0.6179966044142614, "percentage": 61.69, "elapsed_time": "0:12:26", "remaining_time": "0:07:43"}
+{"current_steps": 183, "total_steps": 295, "loss": 0.4613, "lr": 3.85439872551948e-06, "epoch": 0.6213921901528013, "percentage": 62.03, "elapsed_time": "0:12:29", "remaining_time": "0:07:38"}
+{"current_steps": 184, "total_steps": 295, "loss": 0.4489, "lr": 3.796782080920659e-06, "epoch": 0.6247877758913413, "percentage": 62.37, "elapsed_time": "0:12:33", "remaining_time": "0:07:34"}
+{"current_steps": 185, "total_steps": 295, "loss": 0.4433, "lr": 3.7393345377214584e-06, "epoch": 0.6281833616298812, "percentage": 62.71, "elapsed_time": "0:12:36", "remaining_time": "0:07:29"}
+{"current_steps": 186, "total_steps": 295, "loss": 0.4665, "lr": 3.682064169654663e-06, "epoch": 0.631578947368421, "percentage": 63.05, "elapsed_time": "0:12:40", "remaining_time": "0:07:25"}
+{"current_steps": 187, "total_steps": 295, "loss": 0.4363, "lr": 3.6249790255526916e-06, "epoch": 0.634974533106961, "percentage": 63.39, "elapsed_time": "0:12:44", "remaining_time": "0:07:21"}
+{"current_steps": 188, "total_steps": 295, "loss": 0.4646, "lr": 3.568087128216414e-06, "epoch": 0.6383701188455009, "percentage": 63.73, "elapsed_time": "0:12:47", "remaining_time": "0:07:16"}
+{"current_steps": 189, "total_steps": 295, "loss": 0.4571, "lr": 3.511396473287611e-06, "epoch": 0.6417657045840407, "percentage": 64.07, "elapsed_time": "0:12:51", "remaining_time": "0:07:12"}
+{"current_steps": 190, "total_steps": 295, "loss": 0.4484, "lr": 3.4549150281252635e-06, "epoch": 0.6451612903225806, "percentage": 64.41, "elapsed_time": "0:12:54", "remaining_time": "0:07:08"}
+{"current_steps": 191, "total_steps": 295, "loss": 0.457, "lr": 3.398650730685813e-06, "epoch": 0.6485568760611206, "percentage": 64.75, "elapsed_time": "0:12:58", "remaining_time": "0:07:03"}
+{"current_steps": 192, "total_steps": 295, "loss": 0.4467, "lr": 3.3426114884075488e-06, "epoch": 0.6519524617996605, "percentage": 65.08, "elapsed_time": "0:13:03", "remaining_time": "0:07:00"}
+{"current_steps": 193, "total_steps": 295, "loss": 0.4572, "lr": 3.2868051770992935e-06, "epoch": 0.6553480475382003, "percentage": 65.42, "elapsed_time": "0:13:08", "remaining_time": "0:06:56"}
+{"current_steps": 194, "total_steps": 295, "loss": 0.4687, "lr": 3.2312396398335312e-06, "epoch": 0.6587436332767402, "percentage": 65.76, "elapsed_time": "0:13:14", "remaining_time": "0:06:53"}
+{"current_steps": 195, "total_steps": 295, "loss": 0.4336, "lr": 3.175922685844125e-06, "epoch": 0.6621392190152802, "percentage": 66.1, "elapsed_time": "0:13:20", "remaining_time": "0:06:50"}
+{"current_steps": 196, "total_steps": 295, "loss": 0.4486, "lr": 3.1208620894288105e-06, "epoch": 0.66553480475382, "percentage": 66.44, "elapsed_time": "0:13:23", "remaining_time": "0:06:45"}
+{"current_steps": 197, "total_steps": 295, "loss": 0.4588, "lr": 3.0660655888565827e-06, "epoch": 0.6689303904923599, "percentage": 66.78, "elapsed_time": "0:13:28", "remaining_time": "0:06:42"}
+{"current_steps": 198, "total_steps": 295, "loss": 0.4423, "lr": 3.0115408852801535e-06, "epoch": 0.6723259762308998, "percentage": 67.12, "elapsed_time": "0:13:31", "remaining_time": "0:06:37"}
+{"current_steps": 199, "total_steps": 295, "loss": 0.444, "lr": 2.9572956416536267e-06, "epoch": 0.6757215619694398, "percentage": 67.46, "elapsed_time": "0:13:35", "remaining_time": "0:06:33"}
+{"current_steps": 200, "total_steps": 295, "loss": 0.4556, "lr": 2.9033374816555338e-06, "epoch": 0.6791171477079796, "percentage": 67.8, "elapsed_time": "0:13:40", "remaining_time": "0:06:29"}
+{"current_steps": 201, "total_steps": 295, "loss": 0.4727, "lr": 2.8496739886173994e-06, "epoch": 0.6825127334465195, "percentage": 68.14, "elapsed_time": "0:13:44", "remaining_time": "0:06:25"}
+{"current_steps": 202, "total_steps": 295, "loss": 0.432, "lr": 2.7963127044579697e-06, "epoch": 0.6859083191850595, "percentage": 68.47, "elapsed_time": "0:13:47", "remaining_time": "0:06:20"}
+{"current_steps": 203, "total_steps": 295, "loss": 0.4596, "lr": 2.743261128623269e-06, "epoch": 0.6893039049235993, "percentage": 68.81, "elapsed_time": "0:13:51", "remaining_time": "0:06:16"}
+{"current_steps": 204, "total_steps": 295, "loss": 0.4679, "lr": 2.6905267170326143e-06, "epoch": 0.6926994906621392, "percentage": 69.15, "elapsed_time": "0:13:55", "remaining_time": "0:06:12"}
+{"current_steps": 205, "total_steps": 295, "loss": 0.4631, "lr": 2.6381168810307536e-06, "epoch": 0.6960950764006791, "percentage": 69.49, "elapsed_time": "0:14:00", "remaining_time": "0:06:09"}
+{"current_steps": 206, "total_steps": 295, "loss": 0.4475, "lr": 2.5860389863462765e-06, "epoch": 0.6994906621392191, "percentage": 69.83, "elapsed_time": "0:14:04", "remaining_time": "0:06:04"}
+{"current_steps": 207, "total_steps": 295, "loss": 0.4529, "lr": 2.534300352056416e-06, "epoch": 0.7028862478777589, "percentage": 70.17, "elapsed_time": "0:14:10", "remaining_time": "0:06:01"}
+{"current_steps": 208, "total_steps": 295, "loss": 0.4431, "lr": 2.4829082495584244e-06, "epoch": 0.7062818336162988, "percentage": 70.51, "elapsed_time": "0:14:13", "remaining_time": "0:05:57"}
+{"current_steps": 209, "total_steps": 295, "loss": 0.4295, "lr": 2.4318699015476495e-06, "epoch": 0.7096774193548387, "percentage": 70.85, "elapsed_time": "0:14:19", "remaining_time": "0:05:53"}
+{"current_steps": 210, "total_steps": 295, "loss": 0.4462, "lr": 2.3811924810024385e-06, "epoch": 0.7130730050933786, "percentage": 71.19, "elapsed_time": "0:14:23", "remaining_time": "0:05:49"}
+{"current_steps": 211, "total_steps": 295, "loss": 0.4764, "lr": 2.330883110176049e-06, "epoch": 0.7164685908319185, "percentage": 71.53, "elapsed_time": "0:14:26", "remaining_time": "0:05:45"}
+{"current_steps": 212, "total_steps": 295, "loss": 0.4245, "lr": 2.2809488595956746e-06, "epoch": 0.7198641765704584, "percentage": 71.86, "elapsed_time": "0:14:30", "remaining_time": "0:05:40"}
+{"current_steps": 213, "total_steps": 295, "loss": 0.4618, "lr": 2.2313967470687593e-06, "epoch": 0.7232597623089984, "percentage": 72.2, "elapsed_time": "0:14:35", "remaining_time": "0:05:37"}
+{"current_steps": 214, "total_steps": 295, "loss": 0.4614, "lr": 2.18223373669669e-06, "epoch": 0.7266553480475382, "percentage": 72.54, "elapsed_time": "0:14:39", "remaining_time": "0:05:32"}
+{"current_steps": 215, "total_steps": 295, "loss": 0.462, "lr": 2.1334667378960642e-06, "epoch": 0.7300509337860781, "percentage": 72.88, "elapsed_time": "0:14:43", "remaining_time": "0:05:28"}
+{"current_steps": 216, "total_steps": 295, "loss": 0.4634, "lr": 2.0851026044276405e-06, "epoch": 0.733446519524618, "percentage": 73.22, "elapsed_time": "0:14:49", "remaining_time": "0:05:25"}
+{"current_steps": 217, "total_steps": 295, "loss": 0.463, "lr": 2.0371481334330913e-06, "epoch": 0.7368421052631579, "percentage": 73.56, "elapsed_time": "0:14:52", "remaining_time": "0:05:20"}
+{"current_steps": 218, "total_steps": 295, "loss": 0.4788, "lr": 1.9896100644797316e-06, "epoch": 0.7402376910016978, "percentage": 73.9, "elapsed_time": "0:14:56", "remaining_time": "0:05:16"}
+{"current_steps": 219, "total_steps": 295, "loss": 0.4634, "lr": 1.9424950786133414e-06, "epoch": 0.7436332767402377, "percentage": 74.24, "elapsed_time": "0:14:59", "remaining_time": "0:05:12"}
+{"current_steps": 220, "total_steps": 295, "loss": 0.484, "lr": 1.8958097974191909e-06, "epoch": 0.7470288624787776, "percentage": 74.58, "elapsed_time": "0:15:03", "remaining_time": "0:05:08"}
+{"current_steps": 221, "total_steps": 295, "loss": 0.4442, "lr": 1.8495607820914451e-06, "epoch": 0.7504244482173175, "percentage": 74.92, "elapsed_time": "0:15:07", "remaining_time": "0:05:03"}
+{"current_steps": 222, "total_steps": 295, "loss": 0.4241, "lr": 1.8037545325110506e-06, "epoch": 0.7538200339558574, "percentage": 75.25, "elapsed_time": "0:15:12", "remaining_time": "0:05:00"}
+{"current_steps": 223, "total_steps": 295, "loss": 0.4837, "lr": 1.7583974863322272e-06, "epoch": 0.7572156196943973, "percentage": 75.59, "elapsed_time": "0:15:16", "remaining_time": "0:04:55"}
+{"current_steps": 224, "total_steps": 295, "loss": 0.4728, "lr": 1.7134960180777171e-06, "epoch": 0.7606112054329371, "percentage": 75.93, "elapsed_time": "0:15:19", "remaining_time": "0:04:51"}
+{"current_steps": 225, "total_steps": 295, "loss": 0.4301, "lr": 1.6690564382429104e-06, "epoch": 0.7640067911714771, "percentage": 76.27, "elapsed_time": "0:15:25", "remaining_time": "0:04:47"}
+{"current_steps": 226, "total_steps": 295, "loss": 0.4527, "lr": 1.6250849924089485e-06, "epoch": 0.767402376910017, "percentage": 76.61, "elapsed_time": "0:15:29", "remaining_time": "0:04:43"}
+{"current_steps": 227, "total_steps": 295, "loss": 0.4543, "lr": 1.581587860364977e-06, "epoch": 0.7707979626485568, "percentage": 76.95, "elapsed_time": "0:15:32", "remaining_time": "0:04:39"}
+{"current_steps": 228, "total_steps": 295, "loss": 0.4466, "lr": 1.5385711552396227e-06, "epoch": 0.7741935483870968, "percentage": 77.29, "elapsed_time": "0:15:36", "remaining_time": "0:04:35"}
+{"current_steps": 229, "total_steps": 295, "loss": 0.4386, "lr": 1.4960409226418576e-06, "epoch": 0.7775891341256367, "percentage": 77.63, "elapsed_time": "0:15:40", "remaining_time": "0:04:31"}
+{"current_steps": 230, "total_steps": 295, "loss": 0.4533, "lr": 1.4540031398113335e-06, "epoch": 0.7809847198641766, "percentage": 77.97, "elapsed_time": "0:15:44", "remaining_time": "0:04:26"}
+{"current_steps": 231, "total_steps": 295, "loss": 0.4447, "lr": 1.4124637147783431e-06, "epoch": 0.7843803056027164, "percentage": 78.31, "elapsed_time": "0:15:49", "remaining_time": "0:04:22"}
+{"current_steps": 232, "total_steps": 295, "loss": 0.4826, "lr": 1.371428485533498e-06, "epoch": 0.7877758913412564, "percentage": 78.64, "elapsed_time": "0:15:52", "remaining_time": "0:04:18"}
+{"current_steps": 233, "total_steps": 295, "loss": 0.4486, "lr": 1.3309032192072463e-06, "epoch": 0.7911714770797963, "percentage": 78.98, "elapsed_time": "0:15:56", "remaining_time": "0:04:14"}
+{"current_steps": 234, "total_steps": 295, "loss": 0.4396, "lr": 1.29089361125936e-06, "epoch": 0.7945670628183361, "percentage": 79.32, "elapsed_time": "0:16:01", "remaining_time": "0:04:10"}
+{"current_steps": 235, "total_steps": 295, "loss": 0.4666, "lr": 1.251405284678488e-06, "epoch": 0.797962648556876, "percentage": 79.66, "elapsed_time": "0:16:05", "remaining_time": "0:04:06"}
+{"current_steps": 236, "total_steps": 295, "loss": 0.4338, "lr": 1.2124437891918995e-06, "epoch": 0.801358234295416, "percentage": 80.0, "elapsed_time": "0:16:09", "remaining_time": "0:04:02"}
+{"current_steps": 237, "total_steps": 295, "loss": 0.4609, "lr": 1.1740146004855141e-06, "epoch": 0.8047538200339559, "percentage": 80.34, "elapsed_time": "0:16:12", "remaining_time": "0:03:58"}
+{"current_steps": 238, "total_steps": 295, "loss": 0.4505, "lr": 1.1361231194343436e-06, "epoch": 0.8081494057724957, "percentage": 80.68, "elapsed_time": "0:16:16", "remaining_time": "0:03:53"}
+{"current_steps": 239, "total_steps": 295, "loss": 0.4626, "lr": 1.0987746713434578e-06, "epoch": 0.8115449915110357, "percentage": 81.02, "elapsed_time": "0:16:21", "remaining_time": "0:03:49"}
+{"current_steps": 240, "total_steps": 295, "loss": 0.465, "lr": 1.0619745051995473e-06, "epoch": 0.8149405772495756, "percentage": 81.36, "elapsed_time": "0:16:25", "remaining_time": "0:03:45"}
+{"current_steps": 241, "total_steps": 295, "loss": 0.4543, "lr": 1.0257277929332332e-06, "epoch": 0.8183361629881154, "percentage": 81.69, "elapsed_time": "0:16:29", "remaining_time": "0:03:41"}
+{"current_steps": 242, "total_steps": 295, "loss": 0.4676, "lr": 9.900396286922025e-07, "epoch": 0.8217317487266553, "percentage": 82.03, "elapsed_time": "0:16:33", "remaining_time": "0:03:37"}
+{"current_steps": 243, "total_steps": 295, "loss": 0.4366, "lr": 9.549150281252633e-07, "epoch": 0.8251273344651953, "percentage": 82.37, "elapsed_time": "0:16:37", "remaining_time": "0:03:33"}
+{"current_steps": 244, "total_steps": 295, "loss": 0.4404, "lr": 9.203589276774438e-07, "epoch": 0.8285229202037352, "percentage": 82.71, "elapsed_time": "0:16:41", "remaining_time": "0:03:29"}
+{"current_steps": 245, "total_steps": 295, "loss": 0.4447, "lr": 8.86376183896226e-07, "epoch": 0.831918505942275, "percentage": 83.05, "elapsed_time": "0:16:44", "remaining_time": "0:03:25"}
+{"current_steps": 246, "total_steps": 295, "loss": 0.4488, "lr": 8.529715727489912e-07, "epoch": 0.8353140916808149, "percentage": 83.39, "elapsed_time": "0:16:48", "remaining_time": "0:03:20"}
+{"current_steps": 247, "total_steps": 295, "loss": 0.4561, "lr": 8.201497889518073e-07, "epoch": 0.8387096774193549, "percentage": 83.73, "elapsed_time": "0:16:51", "remaining_time": "0:03:16"}
+{"current_steps": 248, "total_steps": 295, "loss": 0.4556, "lr": 7.879154453096305e-07, "epoch": 0.8421052631578947, "percentage": 84.07, "elapsed_time": "0:16:55", "remaining_time": "0:03:12"}
+{"current_steps": 249, "total_steps": 295, "loss": 0.4333, "lr": 7.562730720680111e-07, "epoch": 0.8455008488964346, "percentage": 84.41, "elapsed_time": "0:16:59", "remaining_time": "0:03:08"}
+{"current_steps": 250, "total_steps": 295, "loss": 0.4512, "lr": 7.25227116276413e-07, "epoch": 0.8488964346349746, "percentage": 84.75, "elapsed_time": "0:17:02", "remaining_time": "0:03:04"}
+{"current_steps": 251, "total_steps": 295, "loss": 0.4492, "lr": 6.947819411632223e-07, "epoch": 0.8522920203735145, "percentage": 85.08, "elapsed_time": "0:17:06", "remaining_time": "0:02:59"}
+{"current_steps": 252, "total_steps": 295, "loss": 0.4633, "lr": 6.649418255225298e-07, "epoch": 0.8556876061120543, "percentage": 85.42, "elapsed_time": "0:17:10", "remaining_time": "0:02:55"}
+{"current_steps": 253, "total_steps": 295, "loss": 0.4418, "lr": 6.357109631127889e-07, "epoch": 0.8590831918505942, "percentage": 85.76, "elapsed_time": "0:17:13", "remaining_time": "0:02:51"}
+{"current_steps": 254, "total_steps": 295, "loss": 0.4572, "lr": 6.07093462067419e-07, "epoch": 0.8624787775891342, "percentage": 86.1, "elapsed_time": "0:17:17", "remaining_time": "0:02:47"}
+{"current_steps": 255, "total_steps": 295, "loss": 0.4775, "lr": 5.79093344317449e-07, "epoch": 0.865874363327674, "percentage": 86.44, "elapsed_time": "0:17:20", "remaining_time": "0:02:43"}
+{"current_steps": 256, "total_steps": 295, "loss": 0.4313, "lr": 5.517145450262639e-07, "epoch": 0.8692699490662139, "percentage": 86.78, "elapsed_time": "0:17:27", "remaining_time": "0:02:39"}
+{"current_steps": 257, "total_steps": 295, "loss": 0.4397, "lr": 5.249609120365579e-07, "epoch": 0.8726655348047538, "percentage": 87.12, "elapsed_time": "0:17:31", "remaining_time": "0:02:35"}
+{"current_steps": 258, "total_steps": 295, "loss": 0.4496, "lr": 4.988362053295564e-07, "epoch": 0.8760611205432938, "percentage": 87.46, "elapsed_time": "0:17:35", "remaining_time": "0:02:31"}
+{"current_steps": 259, "total_steps": 295, "loss": 0.4303, "lr": 4.733440964965791e-07, "epoch": 0.8794567062818336, "percentage": 87.8, "elapsed_time": "0:17:40", "remaining_time": "0:02:27"}
+{"current_steps": 260, "total_steps": 295, "loss": 0.4268, "lr": 4.484881682230341e-07, "epoch": 0.8828522920203735, "percentage": 88.14, "elapsed_time": "0:17:43", "remaining_time": "0:02:23"}
+{"current_steps": 261, "total_steps": 295, "loss": 0.4514, "lr": 4.242719137849077e-07, "epoch": 0.8862478777589134, "percentage": 88.47, "elapsed_time": "0:17:47", "remaining_time": "0:02:19"}
+{"current_steps": 262, "total_steps": 295, "loss": 0.4475, "lr": 4.00698736557808e-07, "epoch": 0.8896434634974533, "percentage": 88.81, "elapsed_time": "0:17:52", "remaining_time": "0:02:15"}
+{"current_steps": 263, "total_steps": 295, "loss": 0.4479, "lr": 3.777719495386567e-07, "epoch": 0.8930390492359932, "percentage": 89.15, "elapsed_time": "0:17:56", "remaining_time": "0:02:10"}
+{"current_steps": 264, "total_steps": 295, "loss": 0.4563, "lr": 3.5549477488007853e-07, "epoch": 0.8964346349745331, "percentage": 89.49, "elapsed_time": "0:18:00", "remaining_time": "0:02:06"}
+{"current_steps": 265, "total_steps": 295, "loss": 0.4447, "lr": 3.3387034343755063e-07, "epoch": 0.8998302207130731, "percentage": 89.83, "elapsed_time": "0:18:06", "remaining_time": "0:02:03"}
+{"current_steps": 266, "total_steps": 295, "loss": 0.4416, "lr": 3.1290169432939556e-07, "epoch": 0.9032258064516129, "percentage": 90.17, "elapsed_time": "0:18:10", "remaining_time": "0:01:58"}
+{"current_steps": 267, "total_steps": 295, "loss": 0.4263, "lr": 2.925917745096568e-07, "epoch": 0.9066213921901528, "percentage": 90.51, "elapsed_time": "0:18:14", "remaining_time": "0:01:54"}
+{"current_steps": 268, "total_steps": 295, "loss": 0.4635, "lr": 2.7294343835393366e-07, "epoch": 0.9100169779286927, "percentage": 90.85, "elapsed_time": "0:18:18", "remaining_time": "0:01:50"}
+{"current_steps": 269, "total_steps": 295, "loss": 0.4387, "lr": 2.539594472582213e-07, "epoch": 0.9134125636672326, "percentage": 91.19, "elapsed_time": "0:18:21", "remaining_time": "0:01:46"}
+{"current_steps": 270, "total_steps": 295, "loss": 0.4409, "lr": 2.3564246925082358e-07, "epoch": 0.9168081494057725, "percentage": 91.53, "elapsed_time": "0:18:25", "remaining_time": "0:01:42"}
+{"current_steps": 271, "total_steps": 295, "loss": 0.422, "lr": 2.179950786173879e-07, "epoch": 0.9202037351443124, "percentage": 91.86, "elapsed_time": "0:18:29", "remaining_time": "0:01:38"}
+{"current_steps": 272, "total_steps": 295, "loss": 0.4533, "lr": 2.01019755539108e-07, "epoch": 0.9235993208828522, "percentage": 92.2, "elapsed_time": "0:18:33", "remaining_time": "0:01:34"}
+{"current_steps": 273, "total_steps": 295, "loss": 0.4504, "lr": 1.8471888574415953e-07, "epoch": 0.9269949066213922, "percentage": 92.54, "elapsed_time": "0:18:37", "remaining_time": "0:01:30"}
+{"current_steps": 274, "total_steps": 295, "loss": 0.4616, "lr": 1.690947601724091e-07, "epoch": 0.9303904923599321, "percentage": 92.88, "elapsed_time": "0:18:41", "remaining_time": "0:01:25"}
+{"current_steps": 275, "total_steps": 295, "loss": 0.4362, "lr": 1.5414957465343883e-07, "epoch": 0.933786078098472, "percentage": 93.22, "elapsed_time": "0:18:44", "remaining_time": "0:01:21"}
+{"current_steps": 276, "total_steps": 295, "loss": 0.4476, "lr": 1.3988542959794627e-07, "epoch": 0.9371816638370118, "percentage": 93.56, "elapsed_time": "0:18:50", "remaining_time": "0:01:17"}
+{"current_steps": 277, "total_steps": 295, "loss": 0.4317, "lr": 1.2630432970255014e-07, "epoch": 0.9405772495755518, "percentage": 93.9, "elapsed_time": "0:18:54", "remaining_time": "0:01:13"}
+{"current_steps": 278, "total_steps": 295, "loss": 0.4374, "lr": 1.1340818366804728e-07, "epoch": 0.9439728353140917, "percentage": 94.24, "elapsed_time": "0:18:58", "remaining_time": "0:01:09"}
+{"current_steps": 279, "total_steps": 295, "loss": 0.4463, "lr": 1.0119880393116177e-07, "epoch": 0.9473684210526315, "percentage": 94.58, "elapsed_time": "0:19:01", "remaining_time": "0:01:05"}
+{"current_steps": 280, "total_steps": 295, "loss": 0.4534, "lr": 8.967790640982466e-08, "epoch": 0.9507640067911715, "percentage": 94.92, "elapsed_time": "0:19:05", "remaining_time": "0:01:01"}
+{"current_steps": 281, "total_steps": 295, "loss": 0.4474, "lr": 7.884711026201586e-08, "epoch": 0.9541595925297114, "percentage": 95.25, "elapsed_time": "0:19:09", "remaining_time": "0:00:57"}
+{"current_steps": 282, "total_steps": 295, "loss": 0.4735, "lr": 6.870793765820783e-08, "epoch": 0.9575551782682513, "percentage": 95.59, "elapsed_time": "0:19:13", "remaining_time": "0:00:53"}
+{"current_steps": 283, "total_steps": 295, "loss": 0.4506, "lr": 5.92618135674361e-08, "epoch": 0.9609507640067911, "percentage": 95.93, "elapsed_time": "0:19:17", "remaining_time": "0:00:49"}
+{"current_steps": 284, "total_steps": 295, "loss": 0.4677, "lr": 5.0510065557034526e-08, "epoch": 0.9643463497453311, "percentage": 96.27, "elapsed_time": "0:19:21", "remaining_time": "0:00:45"}
+{"current_steps": 285, "total_steps": 295, "loss": 0.4582, "lr": 4.245392360605727e-08, "epoch": 0.967741935483871, "percentage": 96.61, "elapsed_time": "0:19:27", "remaining_time": "0:00:40"}
+{"current_steps": 286, "total_steps": 295, "loss": 0.4505, "lr": 3.5094519932415417e-08, "epoch": 0.9711375212224108, "percentage": 96.95, "elapsed_time": "0:19:31", "remaining_time": "0:00:36"}
+{"current_steps": 287, "total_steps": 295, "loss": 0.4538, "lr": 2.843288883375539e-08, "epoch": 0.9745331069609507, "percentage": 97.29, "elapsed_time": "0:19:35", "remaining_time": "0:00:32"}
+{"current_steps": 288, "total_steps": 295, "loss": 0.4432, "lr": 2.2469966542096323e-08, "epoch": 0.9779286926994907, "percentage": 97.63, "elapsed_time": "0:19:40", "remaining_time": "0:00:28"}
+{"current_steps": 289, "total_steps": 295, "loss": 0.4567, "lr": 1.7206591092253642e-08, "epoch": 0.9813242784380306, "percentage": 97.97, "elapsed_time": "0:19:43", "remaining_time": "0:00:24"}
+{"current_steps": 290, "total_steps": 295, "loss": 0.4317, "lr": 1.264350220405719e-08, "epoch": 0.9847198641765704, "percentage": 98.31, "elapsed_time": "0:19:48", "remaining_time": "0:00:20"}
+{"current_steps": 291, "total_steps": 295, "loss": 0.4555, "lr": 8.781341178393244e-09, "epoch": 0.9881154499151104, "percentage": 98.64, "elapsed_time": "0:19:52", "remaining_time": "0:00:16"}
+{"current_steps": 292, "total_steps": 295, "loss": 0.4606, "lr": 5.620650807073857e-09, "epoch": 0.9915110356536503, "percentage": 98.98, "elapsed_time": "0:19:55", "remaining_time": "0:00:12"}
+{"current_steps": 293, "total_steps": 295, "loss": 0.4408, "lr": 3.1618752965534295e-09, "epoch": 0.9949066213921901, "percentage": 99.32, "elapsed_time": "0:20:00", "remaining_time": "0:00:08"}
+{"current_steps": 294, "total_steps": 295, "loss": 0.4157, "lr": 1.4053602054991954e-09, "epoch": 0.99830220713073, "percentage": 99.66, "elapsed_time": "0:20:04", "remaining_time": "0:00:04"}
+{"current_steps": 295, "total_steps": 295, "loss": 0.4174, "lr": 3.513523962256349e-10, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "0:20:07", "remaining_time": "0:00:00"}
+{"current_steps": 295, "total_steps": 295, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "0:20:49", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,2108 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 295,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.003395585738539898,
+      "grad_norm": 6.320343624704885,
+      "learning_rate": 0.0,
+      "loss": 0.9899,
+      "step": 1
+    },
+    {
+      "epoch": 0.006791171477079796,
+      "grad_norm": 16.545756923443665,
+      "learning_rate": 3.3333333333333335e-07,
+      "loss": 0.9683,
+      "step": 2
+    },
+    {
+      "epoch": 0.010186757215619695,
+      "grad_norm": 6.124626290117518,
+      "learning_rate": 6.666666666666667e-07,
+      "loss": 0.9722,
+      "step": 3
+    },
+    {
+      "epoch": 0.013582342954159592,
+      "grad_norm": 6.907471369837716,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.9667,
+      "step": 4
+    },
+    {
+      "epoch": 0.01697792869269949,
+      "grad_norm": 6.275431523928528,
+      "learning_rate": 1.3333333333333334e-06,
+      "loss": 0.9814,
+      "step": 5
+    },
+    {
+      "epoch": 0.02037351443123939,
+      "grad_norm": 6.074623009583982,
+      "learning_rate": 1.6666666666666667e-06,
+      "loss": 0.9502,
+      "step": 6
+    },
+    {
+      "epoch": 0.023769100169779286,
+      "grad_norm": 89.99670872123082,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.9411,
+      "step": 7
+    },
+    {
+      "epoch": 0.027164685908319185,
+      "grad_norm": 6.224698090768632,
+      "learning_rate": 2.3333333333333336e-06,
+      "loss": 0.9075,
+      "step": 8
+    },
+    {
+      "epoch": 0.030560271646859084,
+      "grad_norm": 5.913600661145363,
+      "learning_rate": 2.666666666666667e-06,
+      "loss": 0.8979,
+      "step": 9
+    },
+    {
+      "epoch": 0.03395585738539898,
+      "grad_norm": 7.267004601688997,
+      "learning_rate": 3e-06,
+      "loss": 0.8589,
+      "step": 10
+    },
+    {
+      "epoch": 0.03735144312393888,
+      "grad_norm": 4.920293108168373,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 0.8797,
+      "step": 11
+    },
+    {
+      "epoch": 0.04074702886247878,
+      "grad_norm": 7.0251120254603325,
+      "learning_rate": 3.6666666666666666e-06,
+      "loss": 0.785,
+      "step": 12
+    },
+    {
+      "epoch": 0.044142614601018676,
+      "grad_norm": 4.620531235688851,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.7932,
+      "step": 13
+    },
+    {
+      "epoch": 0.04753820033955857,
+      "grad_norm": 8.017042756220283,
+      "learning_rate": 4.333333333333334e-06,
+      "loss": 0.7878,
+      "step": 14
+    },
+    {
+      "epoch": 0.050933786078098474,
+      "grad_norm": 7.168335735527789,
+      "learning_rate": 4.666666666666667e-06,
+      "loss": 0.7854,
+      "step": 15
+    },
+    {
+      "epoch": 0.05432937181663837,
+      "grad_norm": 2.6102783506270137,
+      "learning_rate": 5e-06,
+      "loss": 0.7194,
+      "step": 16
+    },
+    {
+      "epoch": 0.057724957555178265,
+      "grad_norm": 2.8281275726026074,
+      "learning_rate": 5.333333333333334e-06,
+      "loss": 0.7059,
+      "step": 17
+    },
+    {
+      "epoch": 0.06112054329371817,
+      "grad_norm": 2.1330350735122505,
+      "learning_rate": 5.666666666666667e-06,
+      "loss": 0.7236,
+      "step": 18
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 3.3828007198200694,
+      "learning_rate": 6e-06,
+      "loss": 0.7133,
+      "step": 19
+    },
+    {
+      "epoch": 0.06791171477079797,
+      "grad_norm": 1.733407506222324,
+      "learning_rate": 6.333333333333333e-06,
+      "loss": 0.6807,
+      "step": 20
+    },
+    {
+      "epoch": 0.07130730050933787,
+      "grad_norm": 1.5732888077219966,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.6717,
+      "step": 21
+    },
+    {
+      "epoch": 0.07470288624787776,
+      "grad_norm": 2.869294784618442,
+      "learning_rate": 7e-06,
+      "loss": 0.6836,
+      "step": 22
+    },
+    {
+      "epoch": 0.07809847198641766,
+      "grad_norm": 1.539403446338188,
+      "learning_rate": 7.333333333333333e-06,
+      "loss": 0.656,
+      "step": 23
+    },
+    {
+      "epoch": 0.08149405772495756,
+      "grad_norm": 1.390711875015812,
+      "learning_rate": 7.666666666666667e-06,
+      "loss": 0.652,
+      "step": 24
+    },
+    {
+      "epoch": 0.08488964346349745,
+      "grad_norm": 1.4133171062619616,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.6237,
+      "step": 25
+    },
+    {
+      "epoch": 0.08828522920203735,
+      "grad_norm": 1.2675100082185478,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.5999,
+      "step": 26
+    },
+    {
+      "epoch": 0.09168081494057725,
+      "grad_norm": 1.4010718655700412,
+      "learning_rate": 8.666666666666668e-06,
+      "loss": 0.6229,
+      "step": 27
+    },
+    {
+      "epoch": 0.09507640067911714,
+      "grad_norm": 1.4302934299507057,
+      "learning_rate": 9e-06,
+      "loss": 0.5909,
+      "step": 28
+    },
+    {
+      "epoch": 0.09847198641765705,
+      "grad_norm": 1.3417528783040296,
+      "learning_rate": 9.333333333333334e-06,
+      "loss": 0.6236,
+      "step": 29
+    },
+    {
+      "epoch": 0.10186757215619695,
+      "grad_norm": 1.2550326498175997,
+      "learning_rate": 9.666666666666667e-06,
+      "loss": 0.5978,
+      "step": 30
+    },
+    {
+      "epoch": 0.10526315789473684,
+      "grad_norm": 1.2949203936642646,
+      "learning_rate": 1e-05,
+      "loss": 0.6206,
+      "step": 31
+    },
+    {
+      "epoch": 0.10865874363327674,
+      "grad_norm": 1.3404316606873072,
+      "learning_rate": 9.999648647603774e-06,
+      "loss": 0.6103,
+      "step": 32
+    },
+    {
+      "epoch": 0.11205432937181664,
+      "grad_norm": 1.2464543142979883,
+      "learning_rate": 9.998594639794502e-06,
+      "loss": 0.5823,
+      "step": 33
+    },
+    {
+      "epoch": 0.11544991511035653,
+      "grad_norm": 1.28881560865773,
+      "learning_rate": 9.996838124703448e-06,
+      "loss": 0.5724,
+      "step": 34
+    },
+    {
+      "epoch": 0.11884550084889643,
+      "grad_norm": 1.236715524722473,
+      "learning_rate": 9.994379349192927e-06,
+      "loss": 0.6006,
+      "step": 35
+    },
+    {
+      "epoch": 0.12224108658743633,
+      "grad_norm": 1.1748122882399419,
+      "learning_rate": 9.991218658821609e-06,
+      "loss": 0.5834,
+      "step": 36
+    },
+    {
+      "epoch": 0.12563667232597622,
+      "grad_norm": 1.3335055567165142,
+      "learning_rate": 9.987356497795944e-06,
+      "loss": 0.5781,
+      "step": 37
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 1.2573177388462913,
+      "learning_rate": 9.982793408907747e-06,
+      "loss": 0.5613,
+      "step": 38
+    },
+    {
+      "epoch": 0.13242784380305603,
+      "grad_norm": 614.0974853168025,
+      "learning_rate": 9.977530033457906e-06,
+      "loss": 0.5518,
+      "step": 39
+    },
+    {
+      "epoch": 0.13582342954159593,
+      "grad_norm": 1.8130801394796245,
+      "learning_rate": 9.971567111166246e-06,
+      "loss": 0.5875,
+      "step": 40
+    },
+    {
+      "epoch": 0.13921901528013583,
+      "grad_norm": 3.021540974319598,
+      "learning_rate": 9.964905480067585e-06,
+      "loss": 0.5666,
+      "step": 41
+    },
+    {
+      "epoch": 0.14261460101867574,
+      "grad_norm": 1.4307732415515269,
+      "learning_rate": 9.957546076393944e-06,
+      "loss": 0.5829,
+      "step": 42
+    },
+    {
+      "epoch": 0.1460101867572156,
+      "grad_norm": 1.2948953962351142,
+      "learning_rate": 9.949489934442966e-06,
+      "loss": 0.592,
+      "step": 43
+    },
+    {
+      "epoch": 0.1494057724957555,
+      "grad_norm": 1.28013431192408,
+      "learning_rate": 9.940738186432565e-06,
+      "loss": 0.554,
+      "step": 44
+    },
+    {
+      "epoch": 0.15280135823429541,
+      "grad_norm": 1.364283730780204,
+      "learning_rate": 9.931292062341793e-06,
+      "loss": 0.5563,
+      "step": 45
+    },
+    {
+      "epoch": 0.15619694397283532,
+      "grad_norm": 1.1789812327801241,
+      "learning_rate": 9.921152889737985e-06,
+      "loss": 0.5349,
+      "step": 46
+    },
+    {
+      "epoch": 0.15959252971137522,
+      "grad_norm": 1.272995264216437,
+      "learning_rate": 9.910322093590177e-06,
+      "loss": 0.5885,
+      "step": 47
+    },
+    {
+      "epoch": 0.16298811544991512,
+      "grad_norm": 1.2071797315958241,
+      "learning_rate": 9.898801196068839e-06,
+      "loss": 0.555,
+      "step": 48
+    },
+    {
+      "epoch": 0.166383701188455,
+      "grad_norm": 1.2466300266575705,
+      "learning_rate": 9.886591816331953e-06,
+      "loss": 0.5412,
+      "step": 49
+    },
+    {
+      "epoch": 0.1697792869269949,
+      "grad_norm": 1.1951025261533827,
+      "learning_rate": 9.87369567029745e-06,
+      "loss": 0.5266,
+      "step": 50
+    },
+    {
+      "epoch": 0.1731748726655348,
+      "grad_norm": 1.1409578758956818,
+      "learning_rate": 9.860114570402055e-06,
+      "loss": 0.5364,
+      "step": 51
+    },
+    {
+      "epoch": 0.1765704584040747,
+      "grad_norm": 1.194968225105999,
+      "learning_rate": 9.845850425346563e-06,
+      "loss": 0.544,
+      "step": 52
+    },
+    {
+      "epoch": 0.1799660441426146,
+      "grad_norm": 1.105032206844106,
+      "learning_rate": 9.830905239827592e-06,
+      "loss": 0.526,
+      "step": 53
+    },
+    {
+      "epoch": 0.1833616298811545,
+      "grad_norm": 1.0951143452700236,
+      "learning_rate": 9.815281114255841e-06,
+      "loss": 0.5384,
+      "step": 54
+    },
+    {
+      "epoch": 0.1867572156196944,
+      "grad_norm": 1.237628011550819,
+      "learning_rate": 9.798980244460892e-06,
+      "loss": 0.5216,
+      "step": 55
+    },
+    {
+      "epoch": 0.19015280135823429,
+      "grad_norm": 1.2049941993061364,
+      "learning_rate": 9.782004921382612e-06,
+      "loss": 0.5602,
+      "step": 56
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 1.0963401467603866,
+      "learning_rate": 9.764357530749178e-06,
+      "loss": 0.5211,
+      "step": 57
+    },
+    {
+      "epoch": 0.1969439728353141,
+      "grad_norm": 1.1153693044651702,
+      "learning_rate": 9.74604055274178e-06,
+      "loss": 0.5408,
+      "step": 58
+    },
+    {
+      "epoch": 0.200339558573854,
+      "grad_norm": 1.1446372149192978,
+      "learning_rate": 9.727056561646067e-06,
+      "loss": 0.5297,
+      "step": 59
+    },
+    {
+      "epoch": 0.2037351443123939,
+      "grad_norm": 1.1379654178567067,
+      "learning_rate": 9.707408225490343e-06,
+      "loss": 0.5241,
+      "step": 60
+    },
+    {
+      "epoch": 0.2071307300509338,
+      "grad_norm": 1.1133344090884694,
+      "learning_rate": 9.687098305670606e-06,
+      "loss": 0.5373,
+      "step": 61
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 1.14642700050245,
+      "learning_rate": 9.66612965656245e-06,
+      "loss": 0.5397,
+      "step": 62
+    },
+    {
+      "epoch": 0.21392190152801357,
+      "grad_norm": 1.0251599768158084,
+      "learning_rate": 9.644505225119922e-06,
+      "loss": 0.5444,
+      "step": 63
+    },
+    {
+      "epoch": 0.21731748726655348,
+      "grad_norm": 0.9890854380699226,
+      "learning_rate": 9.622228050461345e-06,
+      "loss": 0.4984,
+      "step": 64
+    },
+    {
+      "epoch": 0.22071307300509338,
+      "grad_norm": 1.1117068387014335,
+      "learning_rate": 9.599301263442194e-06,
+      "loss": 0.5214,
+      "step": 65
+    },
+    {
+      "epoch": 0.22410865874363328,
+      "grad_norm": 1.1785090789686485,
+      "learning_rate": 9.575728086215093e-06,
+      "loss": 0.5265,
+      "step": 66
+    },
+    {
+      "epoch": 0.22750424448217318,
+      "grad_norm": 1.0565597242070186,
+      "learning_rate": 9.551511831776966e-06,
+      "loss": 0.5155,
+      "step": 67
+    },
+    {
+      "epoch": 0.23089983022071306,
+      "grad_norm": 1.2278748412937952,
+      "learning_rate": 9.526655903503423e-06,
+      "loss": 0.5277,
+      "step": 68
+    },
+    {
+      "epoch": 0.23429541595925296,
+      "grad_norm": 1.1213279864335683,
+      "learning_rate": 9.501163794670445e-06,
+      "loss": 0.5307,
+      "step": 69
+    },
+    {
+      "epoch": 0.23769100169779286,
+      "grad_norm": 1.1237007845490015,
+      "learning_rate": 9.475039087963443e-06,
+      "loss": 0.51,
+      "step": 70
+    },
+    {
+      "epoch": 0.24108658743633277,
+      "grad_norm": 1.0615774210564253,
+      "learning_rate": 9.448285454973739e-06,
+      "loss": 0.517,
+      "step": 71
+    },
+    {
+      "epoch": 0.24448217317487267,
+      "grad_norm": 1.2104405507373044,
+      "learning_rate": 9.420906655682553e-06,
+      "loss": 0.5223,
+      "step": 72
+    },
+    {
+      "epoch": 0.24787775891341257,
+      "grad_norm": 1.0985265075768098,
+      "learning_rate": 9.392906537932582e-06,
+      "loss": 0.5257,
+      "step": 73
+    },
+    {
+      "epoch": 0.25127334465195245,
+      "grad_norm": 1.0626836434651132,
+      "learning_rate": 9.364289036887214e-06,
+      "loss": 0.5187,
+      "step": 74
+    },
+    {
+      "epoch": 0.2546689303904924,
+      "grad_norm": 1.0234412989238388,
+      "learning_rate": 9.335058174477472e-06,
+      "loss": 0.5127,
+      "step": 75
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 1.0702433549745685,
+      "learning_rate": 9.305218058836778e-06,
+      "loss": 0.4966,
+      "step": 76
+    },
+    {
+      "epoch": 0.2614601018675722,
+      "grad_norm": 1.1228773632449267,
+      "learning_rate": 9.274772883723587e-06,
+      "loss": 0.5265,
+      "step": 77
+    },
+    {
+      "epoch": 0.26485568760611206,
+      "grad_norm": 1.0076068781933833,
+      "learning_rate": 9.24372692793199e-06,
+      "loss": 0.5003,
+      "step": 78
+    },
+    {
+      "epoch": 0.26825127334465193,
+      "grad_norm": 1.0987680518642855,
+      "learning_rate": 9.21208455469037e-06,
+      "loss": 0.4971,
+      "step": 79
+    },
+    {
+      "epoch": 0.27164685908319186,
+      "grad_norm": 1.1051172285171345,
+      "learning_rate": 9.179850211048193e-06,
+      "loss": 0.491,
+      "step": 80
+    },
+    {
+      "epoch": 0.27504244482173174,
+      "grad_norm": 1.0886761114445709,
+      "learning_rate": 9.14702842725101e-06,
+      "loss": 0.5033,
+      "step": 81
+    },
+    {
+      "epoch": 0.27843803056027167,
+      "grad_norm": 1.101989710769037,
+      "learning_rate": 9.113623816103775e-06,
+      "loss": 0.5157,
+      "step": 82
+    },
+    {
+      "epoch": 0.28183361629881154,
+      "grad_norm": 1.0846044794179348,
+      "learning_rate": 9.079641072322555e-06,
+      "loss": 0.5103,
+      "step": 83
+    },
+    {
+      "epoch": 0.28522920203735147,
+      "grad_norm": 1.0994243701550208,
+      "learning_rate": 9.045084971874738e-06,
+      "loss": 0.5118,
+      "step": 84
+    },
+    {
+      "epoch": 0.28862478777589134,
+      "grad_norm": 0.982679989364588,
+      "learning_rate": 9.009960371307798e-06,
+      "loss": 0.4753,
+      "step": 85
+    },
+    {
+      "epoch": 0.2920203735144312,
+      "grad_norm": 1.0974557383394132,
+      "learning_rate": 8.974272207066767e-06,
+      "loss": 0.5065,
+      "step": 86
+    },
+    {
+      "epoch": 0.29541595925297115,
+      "grad_norm": 1.191767026119928,
+      "learning_rate": 8.938025494800454e-06,
+      "loss": 0.496,
+      "step": 87
+    },
+    {
+      "epoch": 0.298811544991511,
+      "grad_norm": 1.0863558052080193,
+      "learning_rate": 8.901225328656543e-06,
+      "loss": 0.4768,
+      "step": 88
+    },
+    {
+      "epoch": 0.30220713073005095,
+      "grad_norm": 1.0292339490206497,
+      "learning_rate": 8.863876880565656e-06,
+      "loss": 0.4982,
+      "step": 89
+    },
+    {
+      "epoch": 0.30560271646859083,
+      "grad_norm": 1.0263488228624327,
+      "learning_rate": 8.825985399514488e-06,
+      "loss": 0.49,
+      "step": 90
+    },
+    {
+      "epoch": 0.3089983022071307,
+      "grad_norm": 1.1083934275101859,
+      "learning_rate": 8.787556210808101e-06,
+      "loss": 0.508,
+      "step": 91
+    },
+    {
+      "epoch": 0.31239388794567063,
+      "grad_norm": 1.0759718773481644,
+      "learning_rate": 8.748594715321512e-06,
+      "loss": 0.493,
+      "step": 92
+    },
+    {
+      "epoch": 0.3157894736842105,
+      "grad_norm": 1.1275176872041444,
+      "learning_rate": 8.70910638874064e-06,
+      "loss": 0.5063,
+      "step": 93
+    },
+    {
+      "epoch": 0.31918505942275044,
+      "grad_norm": 1.0268180450930244,
+      "learning_rate": 8.669096780792754e-06,
+      "loss": 0.4953,
+      "step": 94
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 1.0957503547012442,
+      "learning_rate": 8.628571514466502e-06,
+      "loss": 0.5062,
+      "step": 95
+    },
+    {
+      "epoch": 0.32597623089983024,
+      "grad_norm": 1.0476160480506713,
+      "learning_rate": 8.587536285221656e-06,
+      "loss": 0.5254,
+      "step": 96
+    },
+    {
+      "epoch": 0.3293718166383701,
+      "grad_norm": 0.973467408933069,
+      "learning_rate": 8.545996860188668e-06,
+      "loss": 0.5107,
+      "step": 97
+    },
+    {
+      "epoch": 0.33276740237691,
+      "grad_norm": 0.9690640110532758,
+      "learning_rate": 8.503959077358143e-06,
+      "loss": 0.4767,
+      "step": 98
+    },
+    {
+      "epoch": 0.3361629881154499,
+      "grad_norm": 1.1272513638936046,
+      "learning_rate": 8.46142884476038e-06,
+      "loss": 0.4991,
+      "step": 99
+    },
+    {
+      "epoch": 0.3395585738539898,
+      "grad_norm": 1.089949180068211,
+      "learning_rate": 8.418412139635026e-06,
+      "loss": 0.5202,
+      "step": 100
+    },
+    {
+      "epoch": 0.34295415959252973,
+      "grad_norm": 1.0904976650398113,
+      "learning_rate": 8.374915007591053e-06,
+      "loss": 0.5019,
+      "step": 101
+    },
+    {
+      "epoch": 0.3463497453310696,
+      "grad_norm": 1.0132973869482662,
+      "learning_rate": 8.330943561757092e-06,
+      "loss": 0.4981,
+      "step": 102
+    },
+    {
+      "epoch": 0.34974533106960953,
+      "grad_norm": 1.0144345934272716,
+      "learning_rate": 8.286503981922284e-06,
+      "loss": 0.4828,
+      "step": 103
+    },
+    {
+      "epoch": 0.3531409168081494,
+      "grad_norm": 1.0708364693456955,
+      "learning_rate": 8.241602513667775e-06,
+      "loss": 0.4677,
+      "step": 104
+    },
+    {
+      "epoch": 0.3565365025466893,
+      "grad_norm": 1.0734641865863697,
+      "learning_rate": 8.19624546748895e-06,
+      "loss": 0.4991,
+      "step": 105
+    },
+    {
+      "epoch": 0.3599320882852292,
+      "grad_norm": 1.206730960165882,
+      "learning_rate": 8.150439217908557e-06,
+      "loss": 0.4874,
+      "step": 106
+    },
+    {
+      "epoch": 0.3633276740237691,
+      "grad_norm": 0.990112721589732,
+      "learning_rate": 8.104190202580811e-06,
+      "loss": 0.4613,
+      "step": 107
+    },
+    {
+      "epoch": 0.366723259762309,
+      "grad_norm": 1.135017537331381,
+      "learning_rate": 8.057504921386661e-06,
+      "loss": 0.4981,
+      "step": 108
+    },
+    {
+      "epoch": 0.3701188455008489,
+      "grad_norm": 1.1168906124405507,
+      "learning_rate": 8.010389935520269e-06,
+      "loss": 0.4812,
+      "step": 109
+    },
+    {
+      "epoch": 0.3735144312393888,
+      "grad_norm": 1.109365518761395,
+      "learning_rate": 7.962851866566912e-06,
+      "loss": 0.4901,
+      "step": 110
+    },
+    {
+      "epoch": 0.3769100169779287,
+      "grad_norm": 1.1121750297406146,
+      "learning_rate": 7.914897395572362e-06,
+      "loss": 0.4841,
+      "step": 111
+    },
+    {
+      "epoch": 0.38030560271646857,
+      "grad_norm": 1.10407244769018,
+      "learning_rate": 7.866533262103937e-06,
+      "loss": 0.4813,
+      "step": 112
+    },
+    {
+      "epoch": 0.3837011884550085,
+      "grad_norm": 1.109502100982312,
+      "learning_rate": 7.817766263303312e-06,
+      "loss": 0.4731,
+      "step": 113
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 1.0709239172717544,
+      "learning_rate": 7.768603252931243e-06,
+      "loss": 0.4988,
+      "step": 114
+    },
+    {
+      "epoch": 0.3904923599320883,
+      "grad_norm": 1.0915174076138725,
+      "learning_rate": 7.719051140404327e-06,
+      "loss": 0.4825,
+      "step": 115
+    },
+    {
+      "epoch": 0.3938879456706282,
+      "grad_norm": 1.0965896575304275,
+      "learning_rate": 7.669116889823955e-06,
+      "loss": 0.4742,
+      "step": 116
+    },
+    {
+      "epoch": 0.39728353140916806,
+      "grad_norm": 1.0241307032424805,
+      "learning_rate": 7.6188075189975644e-06,
+      "loss": 0.4743,
+      "step": 117
+    },
+    {
+      "epoch": 0.400679117147708,
+      "grad_norm": 0.9435780818732129,
+      "learning_rate": 7.568130098452352e-06,
+      "loss": 0.4739,
+      "step": 118
+    },
+    {
+      "epoch": 0.40407470288624786,
+      "grad_norm": 1.0248298007190784,
+      "learning_rate": 7.517091750441576e-06,
+      "loss": 0.4814,
+      "step": 119
+    },
+    {
+      "epoch": 0.4074702886247878,
+      "grad_norm": 1.0315243401737155,
+      "learning_rate": 7.465699647943586e-06,
+      "loss": 0.4805,
+      "step": 120
+    },
+    {
+      "epoch": 0.41086587436332767,
+      "grad_norm": 1.091443614674752,
+      "learning_rate": 7.413961013653725e-06,
+      "loss": 0.4953,
+      "step": 121
+    },
+    {
+      "epoch": 0.4142614601018676,
+      "grad_norm": 1.0680561484041602,
+      "learning_rate": 7.361883118969248e-06,
+      "loss": 0.4588,
+      "step": 122
+    },
+    {
+      "epoch": 0.41765704584040747,
+      "grad_norm": 0.9761759048656751,
+      "learning_rate": 7.309473282967387e-06,
+      "loss": 0.469,
+      "step": 123
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 1.0119186737160737,
+      "learning_rate": 7.256738871376733e-06,
+      "loss": 0.4812,
+      "step": 124
+    },
+    {
+      "epoch": 0.4244482173174873,
+      "grad_norm": 0.9670302415957511,
+      "learning_rate": 7.203687295542032e-06,
+      "loss": 0.456,
+      "step": 125
+    },
+    {
+      "epoch": 0.42784380305602715,
+      "grad_norm": 1.001107021910696,
+      "learning_rate": 7.1503260113826035e-06,
+      "loss": 0.4597,
+      "step": 126
+    },
+    {
+      "epoch": 0.4312393887945671,
+      "grad_norm": 1.024857257795839,
+      "learning_rate": 7.09666251834447e-06,
+      "loss": 0.4884,
+      "step": 127
+    },
+    {
+      "epoch": 0.43463497453310695,
+      "grad_norm": 1.0331435407774852,
+      "learning_rate": 7.042704358346375e-06,
+      "loss": 0.4581,
+      "step": 128
+    },
+    {
+      "epoch": 0.4380305602716469,
+      "grad_norm": 1.00793619623988,
+      "learning_rate": 6.988459114719849e-06,
+      "loss": 0.4672,
+      "step": 129
+    },
+    {
+      "epoch": 0.44142614601018676,
+      "grad_norm": 0.9145035341259203,
+      "learning_rate": 6.933934411143419e-06,
+      "loss": 0.4524,
+      "step": 130
+    },
+    {
+      "epoch": 0.44482173174872663,
+      "grad_norm": 1.1118101624400085,
+      "learning_rate": 6.879137910571191e-06,
+      "loss": 0.4738,
+      "step": 131
+    },
+    {
+      "epoch": 0.44821731748726656,
+      "grad_norm": 1.0044078413498474,
+      "learning_rate": 6.824077314155877e-06,
+      "loss": 0.4679,
+      "step": 132
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 1.0568591065810937,
+      "learning_rate": 6.768760360166471e-06,
+      "loss": 0.4909,
+      "step": 133
+    },
+    {
+      "epoch": 0.45500848896434637,
+      "grad_norm": 0.9937372079053678,
+      "learning_rate": 6.713194822900707e-06,
+      "loss": 0.4734,
+      "step": 134
+    },
+    {
+      "epoch": 0.45840407470288624,
+      "grad_norm": 1.0380172124065203,
+      "learning_rate": 6.657388511592453e-06,
+      "loss": 0.4979,
+      "step": 135
+    },
+    {
+      "epoch": 0.4617996604414261,
+      "grad_norm": 0.9710945454163669,
+      "learning_rate": 6.601349269314188e-06,
+      "loss": 0.4826,
+      "step": 136
+    },
+    {
+      "epoch": 0.46519524617996605,
+      "grad_norm": 1.1195229655523649,
+      "learning_rate": 6.545084971874738e-06,
+      "loss": 0.4685,
+      "step": 137
+    },
+    {
+      "epoch": 0.4685908319185059,
+      "grad_norm": 1.079871197310449,
+      "learning_rate": 6.488603526712391e-06,
+      "loss": 0.483,
+      "step": 138
+    },
+    {
+      "epoch": 0.47198641765704585,
+      "grad_norm": 1.0089317127122743,
+      "learning_rate": 6.431912871783587e-06,
+      "loss": 0.5028,
+      "step": 139
+    },
+    {
+      "epoch": 0.47538200339558573,
+      "grad_norm": 0.9671431137430636,
+      "learning_rate": 6.3750209744473105e-06,
+      "loss": 0.4669,
+      "step": 140
+    },
+    {
+      "epoch": 0.47877758913412566,
+      "grad_norm": 1.0520557175009426,
+      "learning_rate": 6.3179358303453386e-06,
+      "loss": 0.4735,
+      "step": 141
+    },
+    {
+      "epoch": 0.48217317487266553,
+      "grad_norm": 1.0538462837406155,
+      "learning_rate": 6.260665462278544e-06,
+      "loss": 0.5051,
+      "step": 142
+    },
+    {
+      "epoch": 0.4855687606112054,
+      "grad_norm": 1.037965055623141,
+      "learning_rate": 6.203217919079343e-06,
+      "loss": 0.4762,
+      "step": 143
+    },
+    {
+      "epoch": 0.48896434634974534,
+      "grad_norm": 1.0176275927771237,
+      "learning_rate": 6.145601274480521e-06,
+      "loss": 0.4869,
+      "step": 144
+    },
+    {
+      "epoch": 0.4923599320882852,
+      "grad_norm": 1.0431340329887582,
+      "learning_rate": 6.08782362598054e-06,
+      "loss": 0.4611,
+      "step": 145
+    },
+    {
+      "epoch": 0.49575551782682514,
+      "grad_norm": 1.0308571103490511,
+      "learning_rate": 6.029893093705492e-06,
+      "loss": 0.4763,
+      "step": 146
+    },
+    {
+      "epoch": 0.499151103565365,
+      "grad_norm": 1.0355702334516321,
+      "learning_rate": 5.971817819267914e-06,
+      "loss": 0.4654,
+      "step": 147
+    },
+    {
+      "epoch": 0.5025466893039049,
+      "grad_norm": 0.9893529623332218,
+      "learning_rate": 5.9136059646225375e-06,
+      "loss": 0.5044,
+      "step": 148
+    },
+    {
+      "epoch": 0.5059422750424448,
+      "grad_norm": 0.960195438000855,
+      "learning_rate": 5.855265710919211e-06,
+      "loss": 0.4435,
+      "step": 149
+    },
+    {
+      "epoch": 0.5093378607809848,
+      "grad_norm": 0.9646163806867624,
+      "learning_rate": 5.796805257353109e-06,
+      "loss": 0.4834,
+      "step": 150
+    },
+    {
+      "epoch": 0.5127334465195246,
+      "grad_norm": 0.9931874900764169,
+      "learning_rate": 5.738232820012407e-06,
+      "loss": 0.461,
+      "step": 151
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 1.0149629817842267,
+      "learning_rate": 5.679556630723592e-06,
+      "loss": 0.4611,
+      "step": 152
+    },
+    {
+      "epoch": 0.5195246179966044,
+      "grad_norm": 1.0022805808479243,
+      "learning_rate": 5.620784935894548e-06,
+      "loss": 0.4712,
+      "step": 153
+    },
+    {
+      "epoch": 0.5229202037351444,
+      "grad_norm": 0.9091219788824919,
+      "learning_rate": 5.561925995355595e-06,
+      "loss": 0.4906,
+      "step": 154
+    },
+    {
+      "epoch": 0.5263157894736842,
+      "grad_norm": 0.9508691466403413,
+      "learning_rate": 5.5029880811986546e-06,
+      "loss": 0.4836,
+      "step": 155
+    },
+    {
+      "epoch": 0.5297113752122241,
+      "grad_norm": 0.9816091460679524,
+      "learning_rate": 5.443979476614674e-06,
+      "loss": 0.4568,
+      "step": 156
+    },
+    {
+      "epoch": 0.533106960950764,
+      "grad_norm": 0.9890714751673821,
+      "learning_rate": 5.384908474729501e-06,
+      "loss": 0.4669,
+      "step": 157
+    },
+    {
+      "epoch": 0.5365025466893039,
+      "grad_norm": 0.9852345937629808,
+      "learning_rate": 5.325783377438357e-06,
+      "loss": 0.4631,
+      "step": 158
+    },
+    {
+      "epoch": 0.5398981324278438,
+      "grad_norm": 0.9992258942689997,
+      "learning_rate": 5.266612494239088e-06,
+      "loss": 0.4824,
+      "step": 159
+    },
+    {
+      "epoch": 0.5432937181663837,
+      "grad_norm": 1.0010167382867965,
+      "learning_rate": 5.207404141064334e-06,
+      "loss": 0.4619,
+      "step": 160
+    },
+    {
+      "epoch": 0.5466893039049237,
+      "grad_norm": 1.117194827109753,
+      "learning_rate": 5.148166639112799e-06,
+      "loss": 0.518,
+      "step": 161
+    },
+    {
+      "epoch": 0.5500848896434635,
+      "grad_norm": 0.9423156223035444,
+      "learning_rate": 5.088908313679788e-06,
+      "loss": 0.4567,
+      "step": 162
+    },
+    {
+      "epoch": 0.5534804753820034,
+      "grad_norm": 0.919109798014551,
+      "learning_rate": 5.029637492987153e-06,
+      "loss": 0.4649,
+      "step": 163
+    },
+    {
+      "epoch": 0.5568760611205433,
+      "grad_norm": 1.1232411363329386,
+      "learning_rate": 4.970362507012848e-06,
+      "loss": 0.4682,
+      "step": 164
+    },
+    {
+      "epoch": 0.5602716468590832,
+      "grad_norm": 0.9560703825281516,
+      "learning_rate": 4.911091686320213e-06,
+      "loss": 0.4594,
+      "step": 165
+    },
+    {
+      "epoch": 0.5636672325976231,
+      "grad_norm": 0.9927617945956294,
+      "learning_rate": 4.8518333608872015e-06,
+      "loss": 0.4266,
+      "step": 166
+    },
+    {
+      "epoch": 0.567062818336163,
+      "grad_norm": 1.0053279255521879,
+      "learning_rate": 4.792595858935668e-06,
+      "loss": 0.4745,
+      "step": 167
+    },
+    {
+      "epoch": 0.5704584040747029,
+      "grad_norm": 1.1247627665383018,
+      "learning_rate": 4.733387505760913e-06,
+      "loss": 0.4863,
+      "step": 168
+    },
+    {
+      "epoch": 0.5738539898132428,
+      "grad_norm": 0.9506838521386066,
+      "learning_rate": 4.674216622561645e-06,
+      "loss": 0.4459,
+      "step": 169
+    },
+    {
+      "epoch": 0.5772495755517827,
+      "grad_norm": 0.8918545360346245,
+      "learning_rate": 4.6150915252705005e-06,
+      "loss": 0.4571,
+      "step": 170
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.9704497436094051,
+      "learning_rate": 4.556020523385326e-06,
+      "loss": 0.4701,
+      "step": 171
+    },
+    {
+      "epoch": 0.5840407470288624,
+      "grad_norm": 0.9830084007827714,
+      "learning_rate": 4.497011918801347e-06,
+      "loss": 0.477,
+      "step": 172
+    },
+    {
+      "epoch": 0.5874363327674024,
+      "grad_norm": 1.0446285595393363,
+      "learning_rate": 4.438074004644407e-06,
+      "loss": 0.4963,
+      "step": 173
+    },
+    {
+      "epoch": 0.5908319185059423,
+      "grad_norm": 0.9430261659822053,
+      "learning_rate": 4.379215064105454e-06,
+      "loss": 0.4502,
+      "step": 174
+    },
+    {
+      "epoch": 0.5942275042444821,
+      "grad_norm": 0.9528761706191039,
+      "learning_rate": 4.32044336927641e-06,
+      "loss": 0.4667,
+      "step": 175
+    },
+    {
+      "epoch": 0.597623089983022,
+      "grad_norm": 0.9712945630496029,
+      "learning_rate": 4.261767179987595e-06,
+      "loss": 0.469,
+      "step": 176
+    },
+    {
+      "epoch": 0.601018675721562,
+      "grad_norm": 0.9203712027671123,
+      "learning_rate": 4.203194742646893e-06,
+      "loss": 0.4549,
+      "step": 177
+    },
+    {
+      "epoch": 0.6044142614601019,
+      "grad_norm": 0.9335750991253596,
+      "learning_rate": 4.1447342890807905e-06,
+      "loss": 0.4589,
+      "step": 178
+    },
+    {
+      "epoch": 0.6078098471986417,
+      "grad_norm": 0.9302840956981335,
+      "learning_rate": 4.086394035377463e-06,
+      "loss": 0.4525,
+      "step": 179
+    },
+    {
+      "epoch": 0.6112054329371817,
+      "grad_norm": 0.9205763667360544,
+      "learning_rate": 4.028182180732088e-06,
+      "loss": 0.4652,
+      "step": 180
+    },
+    {
+      "epoch": 0.6146010186757216,
+      "grad_norm": 0.9053159533350853,
+      "learning_rate": 3.970106906294509e-06,
+      "loss": 0.4887,
+      "step": 181
+    },
+    {
+      "epoch": 0.6179966044142614,
+      "grad_norm": 0.9639072424421392,
+      "learning_rate": 3.912176374019462e-06,
+      "loss": 0.4928,
+      "step": 182
+    },
+    {
+      "epoch": 0.6213921901528013,
+      "grad_norm": 0.8934882136410526,
+      "learning_rate": 3.85439872551948e-06,
+      "loss": 0.4613,
+      "step": 183
+    },
+    {
+      "epoch": 0.6247877758913413,
+      "grad_norm": 0.9142153731395757,
+      "learning_rate": 3.796782080920659e-06,
+      "loss": 0.4489,
+      "step": 184
+    },
+    {
+      "epoch": 0.6281833616298812,
+      "grad_norm": 0.9197655133200423,
+      "learning_rate": 3.7393345377214584e-06,
+      "loss": 0.4433,
+      "step": 185
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.9569197839619695,
+      "learning_rate": 3.682064169654663e-06,
+      "loss": 0.4665,
+      "step": 186
+    },
+    {
+      "epoch": 0.634974533106961,
+      "grad_norm": 0.9362494462879061,
+      "learning_rate": 3.6249790255526916e-06,
+      "loss": 0.4363,
+      "step": 187
+    },
+    {
+      "epoch": 0.6383701188455009,
+      "grad_norm": 0.8671267136185908,
+      "learning_rate": 3.568087128216414e-06,
+      "loss": 0.4646,
+      "step": 188
+    },
+    {
+      "epoch": 0.6417657045840407,
+      "grad_norm": 0.9203019838984255,
+      "learning_rate": 3.511396473287611e-06,
+      "loss": 0.4571,
+      "step": 189
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.9768723974984408,
+      "learning_rate": 3.4549150281252635e-06,
+      "loss": 0.4484,
+      "step": 190
+    },
+    {
+      "epoch": 0.6485568760611206,
+      "grad_norm": 0.9396504959111149,
+      "learning_rate": 3.398650730685813e-06,
+      "loss": 0.457,
+      "step": 191
+    },
+    {
+      "epoch": 0.6519524617996605,
+      "grad_norm": 0.9336335579059931,
+      "learning_rate": 3.3426114884075488e-06,
+      "loss": 0.4467,
+      "step": 192
+    },
+    {
+      "epoch": 0.6553480475382003,
+      "grad_norm": 0.8976643217426382,
+      "learning_rate": 3.2868051770992935e-06,
+      "loss": 0.4572,
+      "step": 193
+    },
+    {
+      "epoch": 0.6587436332767402,
+      "grad_norm": 0.9373033603378723,
+      "learning_rate": 3.2312396398335312e-06,
+      "loss": 0.4687,
+      "step": 194
+    },
+    {
+      "epoch": 0.6621392190152802,
+      "grad_norm": 0.9707294952319975,
+      "learning_rate": 3.175922685844125e-06,
+      "loss": 0.4336,
+      "step": 195
+    },
+    {
+      "epoch": 0.66553480475382,
+      "grad_norm": 0.9583493306758707,
+      "learning_rate": 3.1208620894288105e-06,
+      "loss": 0.4486,
+      "step": 196
+    },
+    {
+      "epoch": 0.6689303904923599,
+      "grad_norm": 0.9820021819293268,
+      "learning_rate": 3.0660655888565827e-06,
+      "loss": 0.4588,
+      "step": 197
+    },
+    {
+      "epoch": 0.6723259762308998,
+      "grad_norm": 0.9400174461623624,
+      "learning_rate": 3.0115408852801535e-06,
+      "loss": 0.4423,
+      "step": 198
+    },
+    {
+      "epoch": 0.6757215619694398,
+      "grad_norm": 0.9603934801371697,
+      "learning_rate": 2.9572956416536267e-06,
+      "loss": 0.444,
+      "step": 199
+    },
+    {
+      "epoch": 0.6791171477079796,
+      "grad_norm": 0.9537870436181752,
+      "learning_rate": 2.9033374816555338e-06,
+      "loss": 0.4556,
+      "step": 200
+    },
+    {
+      "epoch": 0.6825127334465195,
+      "grad_norm": 0.9149135094841083,
+      "learning_rate": 2.8496739886173994e-06,
+      "loss": 0.4727,
+      "step": 201
+    },
+    {
+      "epoch": 0.6859083191850595,
+      "grad_norm": 0.9486316397589806,
+      "learning_rate": 2.7963127044579697e-06,
+      "loss": 0.432,
+      "step": 202
+    },
+    {
+      "epoch": 0.6893039049235993,
+      "grad_norm": 0.9240748880967119,
+      "learning_rate": 2.743261128623269e-06,
+      "loss": 0.4596,
+      "step": 203
+    },
+    {
+      "epoch": 0.6926994906621392,
+      "grad_norm": 0.9573847569611018,
+      "learning_rate": 2.6905267170326143e-06,
+      "loss": 0.4679,
+      "step": 204
+    },
+    {
+      "epoch": 0.6960950764006791,
+      "grad_norm": 0.8845992056166951,
+      "learning_rate": 2.6381168810307536e-06,
+      "loss": 0.4631,
+      "step": 205
+    },
+    {
+      "epoch": 0.6994906621392191,
+      "grad_norm": 1.004797015683845,
+      "learning_rate": 2.5860389863462765e-06,
+      "loss": 0.4475,
+      "step": 206
+    },
+    {
+      "epoch": 0.7028862478777589,
+      "grad_norm": 0.9408081397458115,
+      "learning_rate": 2.534300352056416e-06,
+      "loss": 0.4529,
+      "step": 207
+    },
+    {
+      "epoch": 0.7062818336162988,
+      "grad_norm": 0.8830371296945094,
+      "learning_rate": 2.4829082495584244e-06,
+      "loss": 0.4431,
+      "step": 208
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.918591969840702,
+      "learning_rate": 2.4318699015476495e-06,
+      "loss": 0.4295,
+      "step": 209
+    },
+    {
+      "epoch": 0.7130730050933786,
+      "grad_norm": 0.9126233293580099,
+      "learning_rate": 2.3811924810024385e-06,
+      "loss": 0.4462,
+      "step": 210
+    },
+    {
+      "epoch": 0.7164685908319185,
+      "grad_norm": 0.944049132172904,
+      "learning_rate": 2.330883110176049e-06,
+      "loss": 0.4764,
+      "step": 211
+    },
+    {
+      "epoch": 0.7198641765704584,
+      "grad_norm": 0.8594916774059891,
+      "learning_rate": 2.2809488595956746e-06,
+      "loss": 0.4245,
+      "step": 212
+    },
+    {
+      "epoch": 0.7232597623089984,
+      "grad_norm": 0.9004237430035651,
+      "learning_rate": 2.2313967470687593e-06,
+      "loss": 0.4618,
+      "step": 213
+    },
+    {
+      "epoch": 0.7266553480475382,
+      "grad_norm": 0.9087973453481795,
+      "learning_rate": 2.18223373669669e-06,
+      "loss": 0.4614,
+      "step": 214
+    },
+    {
+      "epoch": 0.7300509337860781,
+      "grad_norm": 0.8919978483526617,
+      "learning_rate": 2.1334667378960642e-06,
+      "loss": 0.462,
+      "step": 215
+    },
+    {
+      "epoch": 0.733446519524618,
+      "grad_norm": 0.9099211098465206,
+      "learning_rate": 2.0851026044276405e-06,
+      "loss": 0.4634,
+      "step": 216
+    },
+    {
+      "epoch": 0.7368421052631579,
+      "grad_norm": 0.8896204484265512,
+      "learning_rate": 2.0371481334330913e-06,
+      "loss": 0.463,
+      "step": 217
+    },
+    {
+      "epoch": 0.7402376910016978,
+      "grad_norm": 0.9519257493559129,
+      "learning_rate": 1.9896100644797316e-06,
+      "loss": 0.4788,
+      "step": 218
+    },
+    {
+      "epoch": 0.7436332767402377,
+      "grad_norm": 0.9116718849117885,
+      "learning_rate": 1.9424950786133414e-06,
+      "loss": 0.4634,
+      "step": 219
+    },
+    {
+      "epoch": 0.7470288624787776,
+      "grad_norm": 0.9263410885293661,
+      "learning_rate": 1.8958097974191909e-06,
+      "loss": 0.484,
+      "step": 220
+    },
+    {
+      "epoch": 0.7504244482173175,
+      "grad_norm": 0.9023645875611066,
+      "learning_rate": 1.8495607820914451e-06,
+      "loss": 0.4442,
+      "step": 221
+    },
+    {
+      "epoch": 0.7538200339558574,
+      "grad_norm": 0.8875288134755263,
+      "learning_rate": 1.8037545325110506e-06,
+      "loss": 0.4241,
+      "step": 222
+    },
+    {
+      "epoch": 0.7572156196943973,
+      "grad_norm": 0.9432820849930534,
+      "learning_rate": 1.7583974863322272e-06,
+      "loss": 0.4837,
+      "step": 223
+    },
+    {
+      "epoch": 0.7606112054329371,
+      "grad_norm": 0.9658199086993665,
+      "learning_rate": 1.7134960180777171e-06,
+      "loss": 0.4728,
+      "step": 224
+    },
+    {
+      "epoch": 0.7640067911714771,
+      "grad_norm": 0.8457866400541412,
+      "learning_rate": 1.6690564382429104e-06,
+      "loss": 0.4301,
+      "step": 225
+    },
+    {
+      "epoch": 0.767402376910017,
+      "grad_norm": 0.8989697214841547,
+      "learning_rate": 1.6250849924089485e-06,
+      "loss": 0.4527,
+      "step": 226
+    },
+    {
+      "epoch": 0.7707979626485568,
+      "grad_norm": 0.9022700419018949,
+      "learning_rate": 1.581587860364977e-06,
+      "loss": 0.4543,
+      "step": 227
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.9173241233268217,
+      "learning_rate": 1.5385711552396227e-06,
+      "loss": 0.4466,
+      "step": 228
+    },
+    {
+      "epoch": 0.7775891341256367,
+      "grad_norm": 0.8617337826425147,
+      "learning_rate": 1.4960409226418576e-06,
+      "loss": 0.4386,
+      "step": 229
+    },
+    {
+      "epoch": 0.7809847198641766,
+      "grad_norm": 0.8625616538339923,
+      "learning_rate": 1.4540031398113335e-06,
+      "loss": 0.4533,
+      "step": 230
+    },
+    {
+      "epoch": 0.7843803056027164,
+      "grad_norm": 0.894048701617508,
+      "learning_rate": 1.4124637147783431e-06,
+      "loss": 0.4447,
+      "step": 231
+    },
+    {
+      "epoch": 0.7877758913412564,
+      "grad_norm": 0.9208844935282854,
+      "learning_rate": 1.371428485533498e-06,
+      "loss": 0.4826,
+      "step": 232
+    },
+    {
+      "epoch": 0.7911714770797963,
+      "grad_norm": 0.8883039726822113,
+      "learning_rate": 1.3309032192072463e-06,
+      "loss": 0.4486,
+      "step": 233
+    },
+    {
+      "epoch": 0.7945670628183361,
+      "grad_norm": 0.899703690304073,
+      "learning_rate": 1.29089361125936e-06,
+      "loss": 0.4396,
+      "step": 234
+    },
+    {
+      "epoch": 0.797962648556876,
+      "grad_norm": 0.942126628969688,
+      "learning_rate": 1.251405284678488e-06,
+      "loss": 0.4666,
+      "step": 235
+    },
+    {
+      "epoch": 0.801358234295416,
+      "grad_norm": 0.876574882692073,
+      "learning_rate": 1.2124437891918995e-06,
+      "loss": 0.4338,
+      "step": 236
+    },
+    {
+      "epoch": 0.8047538200339559,
+      "grad_norm": 0.8738541998828565,
+      "learning_rate": 1.1740146004855141e-06,
+      "loss": 0.4609,
+      "step": 237
+    },
+    {
+      "epoch": 0.8081494057724957,
+      "grad_norm": 0.8732256896169818,
+      "learning_rate": 1.1361231194343436e-06,
+      "loss": 0.4505,
+      "step": 238
+    },
+    {
+      "epoch": 0.8115449915110357,
+      "grad_norm": 0.8634374293174691,
+      "learning_rate": 1.0987746713434578e-06,
+      "loss": 0.4626,
+      "step": 239
+    },
+    {
+      "epoch": 0.8149405772495756,
+      "grad_norm": 0.8709360064531477,
+      "learning_rate": 1.0619745051995473e-06,
+      "loss": 0.465,
+      "step": 240
+    },
+    {
+      "epoch": 0.8183361629881154,
+      "grad_norm": 0.8612129831209381,
+      "learning_rate": 1.0257277929332332e-06,
+      "loss": 0.4543,
+      "step": 241
+    },
+    {
+      "epoch": 0.8217317487266553,
+      "grad_norm": 0.9089176413771565,
+      "learning_rate": 9.900396286922025e-07,
+      "loss": 0.4676,
+      "step": 242
+    },
+    {
+      "epoch": 0.8251273344651953,
+      "grad_norm": 0.8856710278778406,
+      "learning_rate": 9.549150281252633e-07,
+      "loss": 0.4366,
+      "step": 243
+    },
+    {
+      "epoch": 0.8285229202037352,
+      "grad_norm": 0.8188623584178454,
+      "learning_rate": 9.203589276774438e-07,
+      "loss": 0.4404,
+      "step": 244
+    },
+    {
+      "epoch": 0.831918505942275,
+      "grad_norm": 0.9252917982729971,
+      "learning_rate": 8.86376183896226e-07,
+      "loss": 0.4447,
+      "step": 245
+    },
+    {
+      "epoch": 0.8353140916808149,
+      "grad_norm": 0.8784423429363009,
+      "learning_rate": 8.529715727489912e-07,
+      "loss": 0.4488,
+      "step": 246
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.9207047553546756,
+      "learning_rate": 8.201497889518073e-07,
+      "loss": 0.4561,
+      "step": 247
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.886908726068121,
+      "learning_rate": 7.879154453096305e-07,
+      "loss": 0.4556,
+      "step": 248
+    },
+    {
+      "epoch": 0.8455008488964346,
+      "grad_norm": 0.8585705366949494,
+      "learning_rate": 7.562730720680111e-07,
+      "loss": 0.4333,
+      "step": 249
+    },
+    {
+      "epoch": 0.8488964346349746,
+      "grad_norm": 0.8504815527006262,
+      "learning_rate": 7.25227116276413e-07,
+      "loss": 0.4512,
+      "step": 250
+    },
+    {
+      "epoch": 0.8522920203735145,
+      "grad_norm": 0.8201572218259242,
+      "learning_rate": 6.947819411632223e-07,
+      "loss": 0.4492,
+      "step": 251
+    },
+    {
+      "epoch": 0.8556876061120543,
+      "grad_norm": 0.8579475897548395,
+      "learning_rate": 6.649418255225298e-07,
+      "loss": 0.4633,
+      "step": 252
+    },
+    {
+      "epoch": 0.8590831918505942,
+      "grad_norm": 0.8140416489339838,
+      "learning_rate": 6.357109631127889e-07,
+      "loss": 0.4418,
+      "step": 253
+    },
+    {
+      "epoch": 0.8624787775891342,
+      "grad_norm": 0.8552881051968678,
+      "learning_rate": 6.07093462067419e-07,
+      "loss": 0.4572,
+      "step": 254
+    },
+    {
+      "epoch": 0.865874363327674,
+      "grad_norm": 0.8971540169404577,
+      "learning_rate": 5.79093344317449e-07,
+      "loss": 0.4775,
+      "step": 255
+    },
+    {
+      "epoch": 0.8692699490662139,
+      "grad_norm": 0.8058254068183305,
+      "learning_rate": 5.517145450262639e-07,
+      "loss": 0.4313,
+      "step": 256
+    },
+    {
+      "epoch": 0.8726655348047538,
+      "grad_norm": 0.8530487589073922,
+      "learning_rate": 5.249609120365579e-07,
+      "loss": 0.4397,
+      "step": 257
+    },
+    {
+      "epoch": 0.8760611205432938,
+      "grad_norm": 0.9429999718413201,
+      "learning_rate": 4.988362053295564e-07,
+      "loss": 0.4496,
+      "step": 258
+    },
+    {
+      "epoch": 0.8794567062818336,
+      "grad_norm": 0.8375356078246441,
+      "learning_rate": 4.733440964965791e-07,
+      "loss": 0.4303,
+      "step": 259
+    },
+    {
+      "epoch": 0.8828522920203735,
+      "grad_norm": 0.8256703834683652,
+      "learning_rate": 4.484881682230341e-07,
+      "loss": 0.4268,
+      "step": 260
+    },
+    {
+      "epoch": 0.8862478777589134,
+      "grad_norm": 0.8543078561007811,
+      "learning_rate": 4.242719137849077e-07,
+      "loss": 0.4514,
+      "step": 261
+    },
+    {
+      "epoch": 0.8896434634974533,
+      "grad_norm": 0.905414358204414,
+      "learning_rate": 4.00698736557808e-07,
+      "loss": 0.4475,
+      "step": 262
+    },
+    {
+      "epoch": 0.8930390492359932,
+      "grad_norm": 0.8479836834072867,
+      "learning_rate": 3.777719495386567e-07,
+      "loss": 0.4479,
+      "step": 263
+    },
+    {
+      "epoch": 0.8964346349745331,
+      "grad_norm": 0.8646327130622585,
+      "learning_rate": 3.5549477488007853e-07,
+      "loss": 0.4563,
+      "step": 264
+    },
+    {
+      "epoch": 0.8998302207130731,
+      "grad_norm": 0.8363483102479007,
+      "learning_rate": 3.3387034343755063e-07,
+      "loss": 0.4447,
+      "step": 265
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.8772923486186023,
+      "learning_rate": 3.1290169432939556e-07,
+      "loss": 0.4416,
+      "step": 266
+    },
+    {
+      "epoch": 0.9066213921901528,
+      "grad_norm": 0.8320996733730202,
+      "learning_rate": 2.925917745096568e-07,
+      "loss": 0.4263,
+      "step": 267
+    },
+    {
+      "epoch": 0.9100169779286927,
+      "grad_norm": 0.8735391402072182,
+      "learning_rate": 2.7294343835393366e-07,
+      "loss": 0.4635,
+      "step": 268
+    },
+    {
+      "epoch": 0.9134125636672326,
+      "grad_norm": 0.8476560390812548,
+      "learning_rate": 2.539594472582213e-07,
+      "loss": 0.4387,
+      "step": 269
+    },
+    {
+      "epoch": 0.9168081494057725,
+      "grad_norm": 0.8265472830061668,
+      "learning_rate": 2.3564246925082358e-07,
+      "loss": 0.4409,
+      "step": 270
+    },
+    {
+      "epoch": 0.9202037351443124,
+      "grad_norm": 0.8275085794146296,
+      "learning_rate": 2.179950786173879e-07,
+      "loss": 0.422,
+      "step": 271
+    },
+    {
+      "epoch": 0.9235993208828522,
+      "grad_norm": 0.8378368477541204,
+      "learning_rate": 2.01019755539108e-07,
+      "loss": 0.4533,
+      "step": 272
+    },
+    {
+      "epoch": 0.9269949066213922,
+      "grad_norm": 0.8225280576542648,
+      "learning_rate": 1.8471888574415953e-07,
+      "loss": 0.4504,
+      "step": 273
+    },
+    {
+      "epoch": 0.9303904923599321,
+      "grad_norm": 0.8689138287376853,
+      "learning_rate": 1.690947601724091e-07,
+      "loss": 0.4616,
+      "step": 274
+    },
+    {
+      "epoch": 0.933786078098472,
+      "grad_norm": 0.8153963652578823,
+      "learning_rate": 1.5414957465343883e-07,
+      "loss": 0.4362,
+      "step": 275
+    },
+    {
+      "epoch": 0.9371816638370118,
+      "grad_norm": 0.7970051219455544,
+      "learning_rate": 1.3988542959794627e-07,
+      "loss": 0.4476,
+      "step": 276
+    },
+    {
+      "epoch": 0.9405772495755518,
+      "grad_norm": 0.8286263852975556,
+      "learning_rate": 1.2630432970255014e-07,
+      "loss": 0.4317,
+      "step": 277
+    },
+    {
+      "epoch": 0.9439728353140917,
+      "grad_norm": 0.819703548732724,
+      "learning_rate": 1.1340818366804728e-07,
+      "loss": 0.4374,
+      "step": 278
+    },
+    {
+      "epoch": 0.9473684210526315,
+      "grad_norm": 0.8630335285550543,
+      "learning_rate": 1.0119880393116177e-07,
+      "loss": 0.4463,
+      "step": 279
+    },
+    {
+      "epoch": 0.9507640067911715,
+      "grad_norm": 0.8905416016998572,
+      "learning_rate": 8.967790640982466e-08,
+      "loss": 0.4534,
+      "step": 280
+    },
+    {
+      "epoch": 0.9541595925297114,
+      "grad_norm": 0.869553726844365,
+      "learning_rate": 7.884711026201586e-08,
+      "loss": 0.4474,
+      "step": 281
+    },
+    {
+      "epoch": 0.9575551782682513,
+      "grad_norm": 0.8516101851144577,
+      "learning_rate": 6.870793765820783e-08,
+      "loss": 0.4735,
+      "step": 282
+    },
+    {
+      "epoch": 0.9609507640067911,
+      "grad_norm": 0.834166916764426,
+      "learning_rate": 5.92618135674361e-08,
+      "loss": 0.4506,
+      "step": 283
+    },
+    {
+      "epoch": 0.9643463497453311,
+      "grad_norm": 0.8500281237762543,
+      "learning_rate": 5.0510065557034526e-08,
+      "loss": 0.4677,
+      "step": 284
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.833367943844974,
+      "learning_rate": 4.245392360605727e-08,
+      "loss": 0.4582,
+      "step": 285
+    },
+    {
+      "epoch": 0.9711375212224108,
+      "grad_norm": 0.8458107328647104,
+      "learning_rate": 3.5094519932415417e-08,
+      "loss": 0.4505,
+      "step": 286
+    },
+    {
+      "epoch": 0.9745331069609507,
+      "grad_norm": 0.8443895579435003,
+      "learning_rate": 2.843288883375539e-08,
+      "loss": 0.4538,
+      "step": 287
+    },
+    {
+      "epoch": 0.9779286926994907,
+      "grad_norm": 0.8631958241337926,
+      "learning_rate": 2.2469966542096323e-08,
+      "loss": 0.4432,
+      "step": 288
+    },
+    {
+      "epoch": 0.9813242784380306,
+      "grad_norm": 0.886260366562854,
+      "learning_rate": 1.7206591092253642e-08,
+      "loss": 0.4567,
+      "step": 289
+    },
+    {
+      "epoch": 0.9847198641765704,
+      "grad_norm": 0.7994144470255627,
+      "learning_rate": 1.264350220405719e-08,
+      "loss": 0.4317,
+      "step": 290
+    },
+    {
+      "epoch": 0.9881154499151104,
+      "grad_norm": 0.8700681537853684,
+      "learning_rate": 8.781341178393244e-09,
+      "loss": 0.4555,
+      "step": 291
+    },
+    {
+      "epoch": 0.9915110356536503,
+      "grad_norm": 0.8575898812488901,
+      "learning_rate": 5.620650807073857e-09,
+      "loss": 0.4606,
+      "step": 292
+    },
+    {
+      "epoch": 0.9949066213921901,
+      "grad_norm": 0.8616135380601608,
+      "learning_rate": 3.1618752965534295e-09,
+      "loss": 0.4408,
+      "step": 293
+    },
+    {
+      "epoch": 0.99830220713073,
+      "grad_norm": 0.7784658740487929,
+      "learning_rate": 1.4053602054991954e-09,
+      "loss": 0.4157,
+      "step": 294
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.2063313976605035,
+      "learning_rate": 3.513523962256349e-10,
+      "loss": 0.4174,
+      "step": 295
+    },
+    {
+      "epoch": 1.0,
+      "step": 295,
+      "total_flos": 44242544295936.0,
+      "train_loss": 0.5097209928399425,
+      "train_runtime": 1256.4644,
+      "train_samples_per_second": 14.995,
+      "train_steps_per_second": 0.235
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 295,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 2500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 44242544295936.0,
+  "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 sha256:73aa068e9bce41cafed881e595c3959cffc118bd893a4b1b75a37dcc632c76fe
+size 8081