Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,5 @@ 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
+checkpoint-2875/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,60 @@
+---
+library_name: transformers
+license: other
+base_model: Qwen/Qwen2.5-1.5B-Instruct
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: output
+  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. -->
+
+# output
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the TAUR_dev__SIEXP_sft_longmult2d_data__BON__convos 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: 1
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 8
+- total_eval_batch_size: 64
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.05
+- num_epochs: 5
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.52.3
+- Pytorch 2.7.0+cu126
+- Datasets 3.6.0
+- Tokenizers 0.21.1

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": 5.0,
+    "total_flos": 33447808868352.0,
+    "train_loss": 0.42618958398570184,
+    "train_runtime": 1921.5134,
+    "train_samples_per_second": 11.957,
+    "train_steps_per_second": 1.496
+}

chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoint-2875/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
+}

checkpoint-2875/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoint-2875/config.json

@@ -0,0 +1,28 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.52.3",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

checkpoint-2875/generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.52.3"
+}

checkpoint-2875/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6299c5a2b8fb6b4c41b936e5153afbffc4dbeca795cb1e49eccdaaae8d95e486
+size 3087467144

checkpoint-2875/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
+  }
+}

checkpoint-2875/tokenizer.json

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

checkpoint-2875/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
+}

checkpoint-2875/trainer_state.json

@@ -0,0 +1,2043 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 2875,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.017391304347826087,
+      "grad_norm": 3.047056528200788,
+      "learning_rate": 6.25e-07,
+      "loss": 0.6248,
+      "step": 10
+    },
+    {
+      "epoch": 0.034782608695652174,
+      "grad_norm": 2.174143954299078,
+      "learning_rate": 1.3194444444444446e-06,
+      "loss": 0.6183,
+      "step": 20
+    },
+    {
+      "epoch": 0.05217391304347826,
+      "grad_norm": 1.864151160655997,
+      "learning_rate": 2.0138888888888893e-06,
+      "loss": 0.6015,
+      "step": 30
+    },
+    {
+      "epoch": 0.06956521739130435,
+      "grad_norm": 1.667310343818362,
+      "learning_rate": 2.7083333333333334e-06,
+      "loss": 0.5665,
+      "step": 40
+    },
+    {
+      "epoch": 0.08695652173913043,
+      "grad_norm": 2.455941646453557,
+      "learning_rate": 3.4027777777777783e-06,
+      "loss": 0.54,
+      "step": 50
+    },
+    {
+      "epoch": 0.10434782608695652,
+      "grad_norm": 1.9685053486872826,
+      "learning_rate": 4.097222222222222e-06,
+      "loss": 0.5399,
+      "step": 60
+    },
+    {
+      "epoch": 0.12173913043478261,
+      "grad_norm": 2.106997653377029,
+      "learning_rate": 4.791666666666668e-06,
+      "loss": 0.5317,
+      "step": 70
+    },
+    {
+      "epoch": 0.1391304347826087,
+      "grad_norm": 1.780173897349018,
+      "learning_rate": 5.486111111111112e-06,
+      "loss": 0.5183,
+      "step": 80
+    },
+    {
+      "epoch": 0.1565217391304348,
+      "grad_norm": 1.467101178589284,
+      "learning_rate": 6.180555555555557e-06,
+      "loss": 0.5355,
+      "step": 90
+    },
+    {
+      "epoch": 0.17391304347826086,
+      "grad_norm": 1.521288337808483,
+      "learning_rate": 6.875e-06,
+      "loss": 0.5215,
+      "step": 100
+    },
+    {
+      "epoch": 0.19130434782608696,
+      "grad_norm": 1.5769259173207397,
+      "learning_rate": 7.569444444444445e-06,
+      "loss": 0.5239,
+      "step": 110
+    },
+    {
+      "epoch": 0.20869565217391303,
+      "grad_norm": 1.2892203345431497,
+      "learning_rate": 8.263888888888888e-06,
+      "loss": 0.5288,
+      "step": 120
+    },
+    {
+      "epoch": 0.22608695652173913,
+      "grad_norm": 1.6642372146048716,
+      "learning_rate": 8.958333333333334e-06,
+      "loss": 0.5563,
+      "step": 130
+    },
+    {
+      "epoch": 0.24347826086956523,
+      "grad_norm": 1.5134595782170073,
+      "learning_rate": 9.652777777777779e-06,
+      "loss": 0.5047,
+      "step": 140
+    },
+    {
+      "epoch": 0.2608695652173913,
+      "grad_norm": 1.2945158944777713,
+      "learning_rate": 9.999917294352674e-06,
+      "loss": 0.5146,
+      "step": 150
+    },
+    {
+      "epoch": 0.2782608695652174,
+      "grad_norm": 1.7472246575863004,
+      "learning_rate": 9.99925566559051e-06,
+      "loss": 0.5342,
+      "step": 160
+    },
+    {
+      "epoch": 0.2956521739130435,
+      "grad_norm": 1.5193618638564104,
+      "learning_rate": 9.997932495618156e-06,
+      "loss": 0.5083,
+      "step": 170
+    },
+    {
+      "epoch": 0.3130434782608696,
+      "grad_norm": 1.554664459221582,
+      "learning_rate": 9.995947959527968e-06,
+      "loss": 0.5063,
+      "step": 180
+    },
+    {
+      "epoch": 0.33043478260869563,
+      "grad_norm": 2.1482072415475,
+      "learning_rate": 9.993302319929523e-06,
+      "loss": 0.5434,
+      "step": 190
+    },
+    {
+      "epoch": 0.34782608695652173,
+      "grad_norm": 1.4192382000638246,
+      "learning_rate": 9.98999592691486e-06,
+      "loss": 0.5189,
+      "step": 200
+    },
+    {
+      "epoch": 0.3652173913043478,
+      "grad_norm": 1.208599645402354,
+      "learning_rate": 9.986029218012164e-06,
+      "loss": 0.5382,
+      "step": 210
+    },
+    {
+      "epoch": 0.3826086956521739,
+      "grad_norm": 1.1836104891903814,
+      "learning_rate": 9.981402718127853e-06,
+      "loss": 0.5276,
+      "step": 220
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 1.5978655869742624,
+      "learning_rate": 9.976117039477133e-06,
+      "loss": 0.5017,
+      "step": 230
+    },
+    {
+      "epoch": 0.41739130434782606,
+      "grad_norm": 1.3359478565965037,
+      "learning_rate": 9.970172881502982e-06,
+      "loss": 0.4997,
+      "step": 240
+    },
+    {
+      "epoch": 0.43478260869565216,
+      "grad_norm": 1.3948206340232912,
+      "learning_rate": 9.963571030783582e-06,
+      "loss": 0.5582,
+      "step": 250
+    },
+    {
+      "epoch": 0.45217391304347826,
+      "grad_norm": 1.3011147589302579,
+      "learning_rate": 9.956312360928253e-06,
+      "loss": 0.5457,
+      "step": 260
+    },
+    {
+      "epoch": 0.46956521739130436,
+      "grad_norm": 1.080591596133825,
+      "learning_rate": 9.948397832461829e-06,
+      "loss": 0.5552,
+      "step": 270
+    },
+    {
+      "epoch": 0.48695652173913045,
+      "grad_norm": 1.2339442843093948,
+      "learning_rate": 9.93982849269757e-06,
+      "loss": 0.5203,
+      "step": 280
+    },
+    {
+      "epoch": 0.5043478260869565,
+      "grad_norm": 1.5073250471212296,
+      "learning_rate": 9.930605475598566e-06,
+      "loss": 0.5409,
+      "step": 290
+    },
+    {
+      "epoch": 0.5217391304347826,
+      "grad_norm": 1.2705670265704732,
+      "learning_rate": 9.92073000162768e-06,
+      "loss": 0.5404,
+      "step": 300
+    },
+    {
+      "epoch": 0.5391304347826087,
+      "grad_norm": 1.4491195774316425,
+      "learning_rate": 9.910203377586053e-06,
+      "loss": 0.5188,
+      "step": 310
+    },
+    {
+      "epoch": 0.5565217391304348,
+      "grad_norm": 1.4642066613650184,
+      "learning_rate": 9.899026996440173e-06,
+      "loss": 0.4946,
+      "step": 320
+    },
+    {
+      "epoch": 0.5739130434782609,
+      "grad_norm": 1.18880785751321,
+      "learning_rate": 9.887202337137549e-06,
+      "loss": 0.5298,
+      "step": 330
+    },
+    {
+      "epoch": 0.591304347826087,
+      "grad_norm": 1.2818806362850075,
+      "learning_rate": 9.874730964411001e-06,
+      "loss": 0.5299,
+      "step": 340
+    },
+    {
+      "epoch": 0.6086956521739131,
+      "grad_norm": 1.07998471259005,
+      "learning_rate": 9.861614528571607e-06,
+      "loss": 0.5315,
+      "step": 350
+    },
+    {
+      "epoch": 0.6260869565217392,
+      "grad_norm": 1.944925155516315,
+      "learning_rate": 9.847854765290321e-06,
+      "loss": 0.5447,
+      "step": 360
+    },
+    {
+      "epoch": 0.6434782608695652,
+      "grad_norm": 1.511844572828546,
+      "learning_rate": 9.83345349536829e-06,
+      "loss": 0.5291,
+      "step": 370
+    },
+    {
+      "epoch": 0.6608695652173913,
+      "grad_norm": 1.338025282900669,
+      "learning_rate": 9.818412624495911e-06,
+      "loss": 0.5107,
+      "step": 380
+    },
+    {
+      "epoch": 0.6782608695652174,
+      "grad_norm": 1.0620753499324938,
+      "learning_rate": 9.802734143000668e-06,
+      "loss": 0.5441,
+      "step": 390
+    },
+    {
+      "epoch": 0.6956521739130435,
+      "grad_norm": 1.3469601654140646,
+      "learning_rate": 9.786420125583734e-06,
+      "loss": 0.5403,
+      "step": 400
+    },
+    {
+      "epoch": 0.7130434782608696,
+      "grad_norm": 1.1557449160960933,
+      "learning_rate": 9.769472731045451e-06,
+      "loss": 0.5249,
+      "step": 410
+    },
+    {
+      "epoch": 0.7304347826086957,
+      "grad_norm": 1.275996117914776,
+      "learning_rate": 9.751894201999647e-06,
+      "loss": 0.5399,
+      "step": 420
+    },
+    {
+      "epoch": 0.7478260869565218,
+      "grad_norm": 1.2172722683153971,
+      "learning_rate": 9.733686864576883e-06,
+      "loss": 0.5367,
+      "step": 430
+    },
+    {
+      "epoch": 0.7652173913043478,
+      "grad_norm": 1.3421828009403691,
+      "learning_rate": 9.714853128116634e-06,
+      "loss": 0.5251,
+      "step": 440
+    },
+    {
+      "epoch": 0.782608695652174,
+      "grad_norm": 1.0911936537573848,
+      "learning_rate": 9.695395484848476e-06,
+      "loss": 0.555,
+      "step": 450
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 1.3133546561526954,
+      "learning_rate": 9.675316509562282e-06,
+      "loss": 0.5379,
+      "step": 460
+    },
+    {
+      "epoch": 0.8173913043478261,
+      "grad_norm": 1.0450523994543341,
+      "learning_rate": 9.654618859267516e-06,
+      "loss": 0.5115,
+      "step": 470
+    },
+    {
+      "epoch": 0.8347826086956521,
+      "grad_norm": 1.0976625003665517,
+      "learning_rate": 9.633305272841632e-06,
+      "loss": 0.5323,
+      "step": 480
+    },
+    {
+      "epoch": 0.8521739130434782,
+      "grad_norm": 1.1811352794597962,
+      "learning_rate": 9.61137857066764e-06,
+      "loss": 0.5379,
+      "step": 490
+    },
+    {
+      "epoch": 0.8695652173913043,
+      "grad_norm": 1.3489402359600888,
+      "learning_rate": 9.5888416542609e-06,
+      "loss": 0.5353,
+      "step": 500
+    },
+    {
+      "epoch": 0.8869565217391304,
+      "grad_norm": 1.2068307201497395,
+      "learning_rate": 9.565697505885165e-06,
+      "loss": 0.506,
+      "step": 510
+    },
+    {
+      "epoch": 0.9043478260869565,
+      "grad_norm": 1.2234807296718222,
+      "learning_rate": 9.541949188157937e-06,
+      "loss": 0.5291,
+      "step": 520
+    },
+    {
+      "epoch": 0.9217391304347826,
+      "grad_norm": 1.2256177402823996,
+      "learning_rate": 9.517599843645216e-06,
+      "loss": 0.51,
+      "step": 530
+    },
+    {
+      "epoch": 0.9391304347826087,
+      "grad_norm": 1.2281368130382897,
+      "learning_rate": 9.492652694445629e-06,
+      "loss": 0.5136,
+      "step": 540
+    },
+    {
+      "epoch": 0.9565217391304348,
+      "grad_norm": 1.1730852438772608,
+      "learning_rate": 9.467111041764072e-06,
+      "loss": 0.5402,
+      "step": 550
+    },
+    {
+      "epoch": 0.9739130434782609,
+      "grad_norm": 1.0983121942778715,
+      "learning_rate": 9.44097826547486e-06,
+      "loss": 0.5508,
+      "step": 560
+    },
+    {
+      "epoch": 0.991304347826087,
+      "grad_norm": 1.1697621737952535,
+      "learning_rate": 9.414257823674482e-06,
+      "loss": 0.5516,
+      "step": 570
+    },
+    {
+      "epoch": 1.008695652173913,
+      "grad_norm": 1.2095450172211872,
+      "learning_rate": 9.386953252223989e-06,
+      "loss": 0.5011,
+      "step": 580
+    },
+    {
+      "epoch": 1.0260869565217392,
+      "grad_norm": 1.1615174794799035,
+      "learning_rate": 9.35906816428111e-06,
+      "loss": 0.4851,
+      "step": 590
+    },
+    {
+      "epoch": 1.0434782608695652,
+      "grad_norm": 1.0792347526096802,
+      "learning_rate": 9.330606249822125e-06,
+      "loss": 0.4795,
+      "step": 600
+    },
+    {
+      "epoch": 1.0608695652173914,
+      "grad_norm": 1.097575063279078,
+      "learning_rate": 9.30157127515358e-06,
+      "loss": 0.4955,
+      "step": 610
+    },
+    {
+      "epoch": 1.0782608695652174,
+      "grad_norm": 1.1031575053788267,
+      "learning_rate": 9.271967082413899e-06,
+      "loss": 0.4884,
+      "step": 620
+    },
+    {
+      "epoch": 1.0956521739130434,
+      "grad_norm": 1.2767392201135948,
+      "learning_rate": 9.241797589064959e-06,
+      "loss": 0.4655,
+      "step": 630
+    },
+    {
+      "epoch": 1.1130434782608696,
+      "grad_norm": 1.1033082483572434,
+      "learning_rate": 9.211066787373702e-06,
+      "loss": 0.4663,
+      "step": 640
+    },
+    {
+      "epoch": 1.1304347826086956,
+      "grad_norm": 1.576080502703696,
+      "learning_rate": 9.179778743883855e-06,
+      "loss": 0.4725,
+      "step": 650
+    },
+    {
+      "epoch": 1.1478260869565218,
+      "grad_norm": 1.175175499900073,
+      "learning_rate": 9.147937598877797e-06,
+      "loss": 0.4921,
+      "step": 660
+    },
+    {
+      "epoch": 1.1652173913043478,
+      "grad_norm": 1.1370691886093878,
+      "learning_rate": 9.115547565828695e-06,
+      "loss": 0.4649,
+      "step": 670
+    },
+    {
+      "epoch": 1.182608695652174,
+      "grad_norm": 1.2601647572096586,
+      "learning_rate": 9.082612930842942e-06,
+      "loss": 0.5285,
+      "step": 680
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 1.2187851162701455,
+      "learning_rate": 9.049138052092982e-06,
+      "loss": 0.4947,
+      "step": 690
+    },
+    {
+      "epoch": 1.2173913043478262,
+      "grad_norm": 1.1224072986238753,
+      "learning_rate": 9.015127359240603e-06,
+      "loss": 0.4663,
+      "step": 700
+    },
+    {
+      "epoch": 1.2347826086956522,
+      "grad_norm": 1.2366635360634342,
+      "learning_rate": 8.980585352850775e-06,
+      "loss": 0.4828,
+      "step": 710
+    },
+    {
+      "epoch": 1.2521739130434781,
+      "grad_norm": 1.4981539175915906,
+      "learning_rate": 8.94551660379609e-06,
+      "loss": 0.4743,
+      "step": 720
+    },
+    {
+      "epoch": 1.2695652173913043,
+      "grad_norm": 1.0729421363170109,
+      "learning_rate": 8.909925752651914e-06,
+      "loss": 0.4996,
+      "step": 730
+    },
+    {
+      "epoch": 1.2869565217391306,
+      "grad_norm": 1.1955218613972736,
+      "learning_rate": 8.873817509082305e-06,
+      "loss": 0.4863,
+      "step": 740
+    },
+    {
+      "epoch": 1.3043478260869565,
+      "grad_norm": 1.126282527653828,
+      "learning_rate": 8.837196651216802e-06,
+      "loss": 0.465,
+      "step": 750
+    },
+    {
+      "epoch": 1.3217391304347825,
+      "grad_norm": 1.253199532462182,
+      "learning_rate": 8.800068025018133e-06,
+      "loss": 0.5031,
+      "step": 760
+    },
+    {
+      "epoch": 1.3391304347826087,
+      "grad_norm": 1.1627263868301658,
+      "learning_rate": 8.762436543640965e-06,
+      "loss": 0.4783,
+      "step": 770
+    },
+    {
+      "epoch": 1.3565217391304347,
+      "grad_norm": 1.2722684925328063,
+      "learning_rate": 8.724307186781756e-06,
+      "loss": 0.4883,
+      "step": 780
+    },
+    {
+      "epoch": 1.373913043478261,
+      "grad_norm": 1.1113001887408591,
+      "learning_rate": 8.685685000019803e-06,
+      "loss": 0.4947,
+      "step": 790
+    },
+    {
+      "epoch": 1.391304347826087,
+      "grad_norm": 1.0089575387474536,
+      "learning_rate": 8.646575094149568e-06,
+      "loss": 0.4684,
+      "step": 800
+    },
+    {
+      "epoch": 1.4086956521739131,
+      "grad_norm": 1.139961867951673,
+      "learning_rate": 8.606982644504378e-06,
+      "loss": 0.4938,
+      "step": 810
+    },
+    {
+      "epoch": 1.4260869565217391,
+      "grad_norm": 1.099302351381361,
+      "learning_rate": 8.566912890271584e-06,
+      "loss": 0.4723,
+      "step": 820
+    },
+    {
+      "epoch": 1.4434782608695653,
+      "grad_norm": 1.097933896784172,
+      "learning_rate": 8.526371133799277e-06,
+      "loss": 0.5066,
+      "step": 830
+    },
+    {
+      "epoch": 1.4608695652173913,
+      "grad_norm": 1.2653660292097024,
+      "learning_rate": 8.485362739894617e-06,
+      "loss": 0.4809,
+      "step": 840
+    },
+    {
+      "epoch": 1.4782608695652173,
+      "grad_norm": 1.031416554665846,
+      "learning_rate": 8.443893135113956e-06,
+      "loss": 0.4849,
+      "step": 850
+    },
+    {
+      "epoch": 1.4956521739130435,
+      "grad_norm": 1.2133975894170272,
+      "learning_rate": 8.401967807044713e-06,
+      "loss": 0.4731,
+      "step": 860
+    },
+    {
+      "epoch": 1.5130434782608697,
+      "grad_norm": 1.3527575598259336,
+      "learning_rate": 8.359592303579241e-06,
+      "loss": 0.4993,
+      "step": 870
+    },
+    {
+      "epoch": 1.5304347826086957,
+      "grad_norm": 1.5538636433489414,
+      "learning_rate": 8.316772232180677e-06,
+      "loss": 0.4831,
+      "step": 880
+    },
+    {
+      "epoch": 1.5478260869565217,
+      "grad_norm": 1.0668506396404902,
+      "learning_rate": 8.273513259140911e-06,
+      "loss": 0.4549,
+      "step": 890
+    },
+    {
+      "epoch": 1.5652173913043477,
+      "grad_norm": 1.3545786110485585,
+      "learning_rate": 8.22982110883079e-06,
+      "loss": 0.4889,
+      "step": 900
+    },
+    {
+      "epoch": 1.5826086956521739,
+      "grad_norm": 1.343041115784667,
+      "learning_rate": 8.185701562942614e-06,
+      "loss": 0.507,
+      "step": 910
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 1.2218861872968096,
+      "learning_rate": 8.141160459725063e-06,
+      "loss": 0.4812,
+      "step": 920
+    },
+    {
+      "epoch": 1.617391304347826,
+      "grad_norm": 1.2015960018363139,
+      "learning_rate": 8.096203693210626e-06,
+      "loss": 0.4965,
+      "step": 930
+    },
+    {
+      "epoch": 1.634782608695652,
+      "grad_norm": 1.096282312846291,
+      "learning_rate": 8.050837212435662e-06,
+      "loss": 0.4557,
+      "step": 940
+    },
+    {
+      "epoch": 1.6521739130434783,
+      "grad_norm": 1.0909506176288037,
+      "learning_rate": 8.00506702065318e-06,
+      "loss": 0.4702,
+      "step": 950
+    },
+    {
+      "epoch": 1.6695652173913045,
+      "grad_norm": 1.5020729724704236,
+      "learning_rate": 7.958899174538423e-06,
+      "loss": 0.4695,
+      "step": 960
+    },
+    {
+      "epoch": 1.6869565217391305,
+      "grad_norm": 1.1184823557537737,
+      "learning_rate": 7.912339783387429e-06,
+      "loss": 0.4869,
+      "step": 970
+    },
+    {
+      "epoch": 1.7043478260869565,
+      "grad_norm": 1.5926675735162503,
+      "learning_rate": 7.865395008308572e-06,
+      "loss": 0.4556,
+      "step": 980
+    },
+    {
+      "epoch": 1.7217391304347827,
+      "grad_norm": 1.357255668450024,
+      "learning_rate": 7.818071061407295e-06,
+      "loss": 0.5226,
+      "step": 990
+    },
+    {
+      "epoch": 1.7391304347826086,
+      "grad_norm": 1.1738545925108272,
+      "learning_rate": 7.770374204964062e-06,
+      "loss": 0.4651,
+      "step": 1000
+    },
+    {
+      "epoch": 1.7565217391304349,
+      "grad_norm": 1.1820022034633777,
+      "learning_rate": 7.722310750605693e-06,
+      "loss": 0.498,
+      "step": 1010
+    },
+    {
+      "epoch": 1.7739130434782608,
+      "grad_norm": 1.3439499575644294,
+      "learning_rate": 7.673887058470155e-06,
+      "loss": 0.4832,
+      "step": 1020
+    },
+    {
+      "epoch": 1.7913043478260868,
+      "grad_norm": 1.5304830271309717,
+      "learning_rate": 7.625109536364938e-06,
+      "loss": 0.4716,
+      "step": 1030
+    },
+    {
+      "epoch": 1.808695652173913,
+      "grad_norm": 1.2364034202681973,
+      "learning_rate": 7.5759846389191234e-06,
+      "loss": 0.4696,
+      "step": 1040
+    },
+    {
+      "epoch": 1.8260869565217392,
+      "grad_norm": 1.0229154001597585,
+      "learning_rate": 7.526518866729256e-06,
+      "loss": 0.5235,
+      "step": 1050
+    },
+    {
+      "epoch": 1.8434782608695652,
+      "grad_norm": 1.7878383661619177,
+      "learning_rate": 7.476718765499131e-06,
+      "loss": 0.4936,
+      "step": 1060
+    },
+    {
+      "epoch": 1.8608695652173912,
+      "grad_norm": 1.1220136346350835,
+      "learning_rate": 7.426590925173624e-06,
+      "loss": 0.4954,
+      "step": 1070
+    },
+    {
+      "epoch": 1.8782608695652174,
+      "grad_norm": 1.0276108610586765,
+      "learning_rate": 7.376141979066644e-06,
+      "loss": 0.5226,
+      "step": 1080
+    },
+    {
+      "epoch": 1.8956521739130436,
+      "grad_norm": 1.0538696180373945,
+      "learning_rate": 7.325378602983371e-06,
+      "loss": 0.4997,
+      "step": 1090
+    },
+    {
+      "epoch": 1.9130434782608696,
+      "grad_norm": 1.0941944221450914,
+      "learning_rate": 7.274307514336854e-06,
+      "loss": 0.5069,
+      "step": 1100
+    },
+    {
+      "epoch": 1.9304347826086956,
+      "grad_norm": 1.0475459053099843,
+      "learning_rate": 7.222935471259113e-06,
+      "loss": 0.4516,
+      "step": 1110
+    },
+    {
+      "epoch": 1.9478260869565216,
+      "grad_norm": 0.9438781712939313,
+      "learning_rate": 7.171269271706847e-06,
+      "loss": 0.4771,
+      "step": 1120
+    },
+    {
+      "epoch": 1.9652173913043478,
+      "grad_norm": 1.26652793242089,
+      "learning_rate": 7.119315752561879e-06,
+      "loss": 0.4628,
+      "step": 1130
+    },
+    {
+      "epoch": 1.982608695652174,
+      "grad_norm": 1.0230830545480807,
+      "learning_rate": 7.0670817887264375e-06,
+      "loss": 0.4961,
+      "step": 1140
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.3043673360211918,
+      "learning_rate": 7.014574292213423e-06,
+      "loss": 0.4854,
+      "step": 1150
+    },
+    {
+      "epoch": 2.017391304347826,
+      "grad_norm": 1.2839938386528644,
+      "learning_rate": 6.9618002112317525e-06,
+      "loss": 0.4159,
+      "step": 1160
+    },
+    {
+      "epoch": 2.034782608695652,
+      "grad_norm": 1.485302697617233,
+      "learning_rate": 6.908766529266915e-06,
+      "loss": 0.4513,
+      "step": 1170
+    },
+    {
+      "epoch": 2.0521739130434784,
+      "grad_norm": 1.3513626513686101,
+      "learning_rate": 6.855480264156864e-06,
+      "loss": 0.4147,
+      "step": 1180
+    },
+    {
+      "epoch": 2.0695652173913044,
+      "grad_norm": 1.1971012611734106,
+      "learning_rate": 6.8019484671633586e-06,
+      "loss": 0.4279,
+      "step": 1190
+    },
+    {
+      "epoch": 2.0869565217391304,
+      "grad_norm": 1.1493410270586013,
+      "learning_rate": 6.748178222038891e-06,
+      "loss": 0.4121,
+      "step": 1200
+    },
+    {
+      "epoch": 2.1043478260869564,
+      "grad_norm": 1.2061559921790461,
+      "learning_rate": 6.694176644089297e-06,
+      "loss": 0.4302,
+      "step": 1210
+    },
+    {
+      "epoch": 2.121739130434783,
+      "grad_norm": 1.2134461208204077,
+      "learning_rate": 6.639950879232218e-06,
+      "loss": 0.4143,
+      "step": 1220
+    },
+    {
+      "epoch": 2.139130434782609,
+      "grad_norm": 1.1936990104884397,
+      "learning_rate": 6.585508103051478e-06,
+      "loss": 0.411,
+      "step": 1230
+    },
+    {
+      "epoch": 2.1565217391304348,
+      "grad_norm": 1.473818496639146,
+      "learning_rate": 6.530855519847568e-06,
+      "loss": 0.4247,
+      "step": 1240
+    },
+    {
+      "epoch": 2.1739130434782608,
+      "grad_norm": 1.2273898491461919,
+      "learning_rate": 6.4760003616843106e-06,
+      "loss": 0.4,
+      "step": 1250
+    },
+    {
+      "epoch": 2.1913043478260867,
+      "grad_norm": 1.193058454958422,
+      "learning_rate": 6.420949887431855e-06,
+      "loss": 0.4016,
+      "step": 1260
+    },
+    {
+      "epoch": 2.208695652173913,
+      "grad_norm": 1.3961066733797438,
+      "learning_rate": 6.3657113818061225e-06,
+      "loss": 0.4408,
+      "step": 1270
+    },
+    {
+      "epoch": 2.226086956521739,
+      "grad_norm": 1.292360770899796,
+      "learning_rate": 6.310292154404844e-06,
+      "loss": 0.3999,
+      "step": 1280
+    },
+    {
+      "epoch": 2.243478260869565,
+      "grad_norm": 3.4982840024869417,
+      "learning_rate": 6.254699538740292e-06,
+      "loss": 0.4329,
+      "step": 1290
+    },
+    {
+      "epoch": 2.260869565217391,
+      "grad_norm": 2.265019151887774,
+      "learning_rate": 6.198940891268844e-06,
+      "loss": 0.4282,
+      "step": 1300
+    },
+    {
+      "epoch": 2.2782608695652176,
+      "grad_norm": 1.1398620538733453,
+      "learning_rate": 6.14302359041753e-06,
+      "loss": 0.4222,
+      "step": 1310
+    },
+    {
+      "epoch": 2.2956521739130435,
+      "grad_norm": 1.523968350586471,
+      "learning_rate": 6.086955035607655e-06,
+      "loss": 0.4317,
+      "step": 1320
+    },
+    {
+      "epoch": 2.3130434782608695,
+      "grad_norm": 2.2066066593878197,
+      "learning_rate": 6.030742646275647e-06,
+      "loss": 0.4212,
+      "step": 1330
+    },
+    {
+      "epoch": 2.3304347826086955,
+      "grad_norm": 1.2410016732543188,
+      "learning_rate": 5.9743938608912626e-06,
+      "loss": 0.4334,
+      "step": 1340
+    },
+    {
+      "epoch": 2.3478260869565215,
+      "grad_norm": 1.210595936193342,
+      "learning_rate": 5.917916135973263e-06,
+      "loss": 0.4153,
+      "step": 1350
+    },
+    {
+      "epoch": 2.365217391304348,
+      "grad_norm": 1.4742810962432356,
+      "learning_rate": 5.861316945102717e-06,
+      "loss": 0.4103,
+      "step": 1360
+    },
+    {
+      "epoch": 2.382608695652174,
+      "grad_norm": 1.1807404650454025,
+      "learning_rate": 5.804603777934032e-06,
+      "loss": 0.4104,
+      "step": 1370
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 1.32289460697371,
+      "learning_rate": 5.74778413920386e-06,
+      "loss": 0.4245,
+      "step": 1380
+    },
+    {
+      "epoch": 2.417391304347826,
+      "grad_norm": 1.4542008533192958,
+      "learning_rate": 5.690865547738021e-06,
+      "loss": 0.4412,
+      "step": 1390
+    },
+    {
+      "epoch": 2.4347826086956523,
+      "grad_norm": 1.2739934708683387,
+      "learning_rate": 5.6338555354565445e-06,
+      "loss": 0.4281,
+      "step": 1400
+    },
+    {
+      "epoch": 2.4521739130434783,
+      "grad_norm": 1.563662237308135,
+      "learning_rate": 5.576761646376987e-06,
+      "loss": 0.4359,
+      "step": 1410
+    },
+    {
+      "epoch": 2.4695652173913043,
+      "grad_norm": 1.225022509026475,
+      "learning_rate": 5.519591435616153e-06,
+      "loss": 0.4377,
+      "step": 1420
+    },
+    {
+      "epoch": 2.4869565217391303,
+      "grad_norm": 1.1686568195887326,
+      "learning_rate": 5.462352468390333e-06,
+      "loss": 0.4229,
+      "step": 1430
+    },
+    {
+      "epoch": 2.5043478260869563,
+      "grad_norm": 1.2575917217265098,
+      "learning_rate": 5.405052319014223e-06,
+      "loss": 0.4225,
+      "step": 1440
+    },
+    {
+      "epoch": 2.5217391304347827,
+      "grad_norm": 1.2880775613395745,
+      "learning_rate": 5.347698569898624e-06,
+      "loss": 0.4245,
+      "step": 1450
+    },
+    {
+      "epoch": 2.5391304347826087,
+      "grad_norm": 1.3156168488856514,
+      "learning_rate": 5.290298810547083e-06,
+      "loss": 0.3993,
+      "step": 1460
+    },
+    {
+      "epoch": 2.5565217391304347,
+      "grad_norm": 1.1736379125037195,
+      "learning_rate": 5.232860636551583e-06,
+      "loss": 0.4324,
+      "step": 1470
+    },
+    {
+      "epoch": 2.573913043478261,
+      "grad_norm": 1.2353830078230719,
+      "learning_rate": 5.175391648587443e-06,
+      "loss": 0.4185,
+      "step": 1480
+    },
+    {
+      "epoch": 2.591304347826087,
+      "grad_norm": 1.1458194395748447,
+      "learning_rate": 5.117899451407526e-06,
+      "loss": 0.4309,
+      "step": 1490
+    },
+    {
+      "epoch": 2.608695652173913,
+      "grad_norm": 1.2538290106691092,
+      "learning_rate": 5.060391652835925e-06,
+      "loss": 0.4292,
+      "step": 1500
+    },
+    {
+      "epoch": 2.626086956521739,
+      "grad_norm": 1.571430145629218,
+      "learning_rate": 5.002875862761234e-06,
+      "loss": 0.4268,
+      "step": 1510
+    },
+    {
+      "epoch": 2.643478260869565,
+      "grad_norm": 1.1840870452769912,
+      "learning_rate": 4.9453596921295435e-06,
+      "loss": 0.4319,
+      "step": 1520
+    },
+    {
+      "epoch": 2.660869565217391,
+      "grad_norm": 1.0371855886325287,
+      "learning_rate": 4.8878507519373055e-06,
+      "loss": 0.4378,
+      "step": 1530
+    },
+    {
+      "epoch": 2.6782608695652175,
+      "grad_norm": 1.0983294473272651,
+      "learning_rate": 4.830356652224181e-06,
+      "loss": 0.4058,
+      "step": 1540
+    },
+    {
+      "epoch": 2.6956521739130435,
+      "grad_norm": 1.127081289246426,
+      "learning_rate": 4.77288500106602e-06,
+      "loss": 0.4165,
+      "step": 1550
+    },
+    {
+      "epoch": 2.7130434782608694,
+      "grad_norm": 1.236794619724129,
+      "learning_rate": 4.715443403568103e-06,
+      "loss": 0.4148,
+      "step": 1560
+    },
+    {
+      "epoch": 2.730434782608696,
+      "grad_norm": 1.5345488197083625,
+      "learning_rate": 4.65803946085877e-06,
+      "loss": 0.4088,
+      "step": 1570
+    },
+    {
+      "epoch": 2.747826086956522,
+      "grad_norm": 1.3004765992472387,
+      "learning_rate": 4.600680769083585e-06,
+      "loss": 0.4359,
+      "step": 1580
+    },
+    {
+      "epoch": 2.765217391304348,
+      "grad_norm": 1.2501644106321015,
+      "learning_rate": 4.543374918400142e-06,
+      "loss": 0.4386,
+      "step": 1590
+    },
+    {
+      "epoch": 2.782608695652174,
+      "grad_norm": 1.3823645221856542,
+      "learning_rate": 4.486129491973687e-06,
+      "loss": 0.4517,
+      "step": 1600
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 1.2761252508293937,
+      "learning_rate": 4.4289520649736475e-06,
+      "loss": 0.4283,
+      "step": 1610
+    },
+    {
+      "epoch": 2.8173913043478263,
+      "grad_norm": 1.1632722443857555,
+      "learning_rate": 4.371850203571225e-06,
+      "loss": 0.421,
+      "step": 1620
+    },
+    {
+      "epoch": 2.8347826086956522,
+      "grad_norm": 1.2193532972111443,
+      "learning_rate": 4.314831463938184e-06,
+      "loss": 0.4188,
+      "step": 1630
+    },
+    {
+      "epoch": 2.8521739130434782,
+      "grad_norm": 1.4424575510136903,
+      "learning_rate": 4.257903391246954e-06,
+      "loss": 0.4094,
+      "step": 1640
+    },
+    {
+      "epoch": 2.869565217391304,
+      "grad_norm": 1.095166597880803,
+      "learning_rate": 4.201073518672195e-06,
+      "loss": 0.4411,
+      "step": 1650
+    },
+    {
+      "epoch": 2.8869565217391306,
+      "grad_norm": 1.2662835313537542,
+      "learning_rate": 4.144349366393949e-06,
+      "loss": 0.4423,
+      "step": 1660
+    },
+    {
+      "epoch": 2.9043478260869566,
+      "grad_norm": 1.1922956806962224,
+      "learning_rate": 4.08773844060251e-06,
+      "loss": 0.4254,
+      "step": 1670
+    },
+    {
+      "epoch": 2.9217391304347826,
+      "grad_norm": 1.257711038107532,
+      "learning_rate": 4.031248232505139e-06,
+      "loss": 0.4358,
+      "step": 1680
+    },
+    {
+      "epoch": 2.9391304347826086,
+      "grad_norm": 1.6971639546827986,
+      "learning_rate": 3.97488621733478e-06,
+      "loss": 0.4084,
+      "step": 1690
+    },
+    {
+      "epoch": 2.9565217391304346,
+      "grad_norm": 1.349596141789289,
+      "learning_rate": 3.918659853360864e-06,
+      "loss": 0.4274,
+      "step": 1700
+    },
+    {
+      "epoch": 2.973913043478261,
+      "grad_norm": 1.3678548460193103,
+      "learning_rate": 3.862576580902383e-06,
+      "loss": 0.4187,
+      "step": 1710
+    },
+    {
+      "epoch": 2.991304347826087,
+      "grad_norm": 1.2860411793928146,
+      "learning_rate": 3.8066438213433234e-06,
+      "loss": 0.4303,
+      "step": 1720
+    },
+    {
+      "epoch": 3.008695652173913,
+      "grad_norm": 1.5134066645346198,
+      "learning_rate": 3.7508689761506055e-06,
+      "loss": 0.3752,
+      "step": 1730
+    },
+    {
+      "epoch": 3.026086956521739,
+      "grad_norm": 1.424990341239896,
+      "learning_rate": 3.6952594258946693e-06,
+      "loss": 0.3434,
+      "step": 1740
+    },
+    {
+      "epoch": 3.0434782608695654,
+      "grad_norm": 1.6649951275575265,
+      "learning_rate": 3.6398225292728185e-06,
+      "loss": 0.3607,
+      "step": 1750
+    },
+    {
+      "epoch": 3.0608695652173914,
+      "grad_norm": 1.8127043018682598,
+      "learning_rate": 3.584565622135453e-06,
+      "loss": 0.3623,
+      "step": 1760
+    },
+    {
+      "epoch": 3.0782608695652174,
+      "grad_norm": 1.3226724796309823,
+      "learning_rate": 3.5294960165153363e-06,
+      "loss": 0.3579,
+      "step": 1770
+    },
+    {
+      "epoch": 3.0956521739130434,
+      "grad_norm": 1.4747389958766943,
+      "learning_rate": 3.474620999660007e-06,
+      "loss": 0.3973,
+      "step": 1780
+    },
+    {
+      "epoch": 3.1130434782608694,
+      "grad_norm": 1.9045203788068308,
+      "learning_rate": 3.4199478330674745e-06,
+      "loss": 0.3579,
+      "step": 1790
+    },
+    {
+      "epoch": 3.130434782608696,
+      "grad_norm": 1.2899295921725031,
+      "learning_rate": 3.365483751525317e-06,
+      "loss": 0.3634,
+      "step": 1800
+    },
+    {
+      "epoch": 3.1478260869565218,
+      "grad_norm": 1.4141207800877191,
+      "learning_rate": 3.3112359621533193e-06,
+      "loss": 0.3427,
+      "step": 1810
+    },
+    {
+      "epoch": 3.1652173913043478,
+      "grad_norm": 1.4287974659319675,
+      "learning_rate": 3.257211643449768e-06,
+      "loss": 0.3425,
+      "step": 1820
+    },
+    {
+      "epoch": 3.1826086956521737,
+      "grad_norm": 1.3872599954983553,
+      "learning_rate": 3.203417944341536e-06,
+      "loss": 0.3458,
+      "step": 1830
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 2.2293598598474245,
+      "learning_rate": 3.149861983238082e-06,
+      "loss": 0.3689,
+      "step": 1840
+    },
+    {
+      "epoch": 3.217391304347826,
+      "grad_norm": 2.0325771541464217,
+      "learning_rate": 3.0965508470894812e-06,
+      "loss": 0.3521,
+      "step": 1850
+    },
+    {
+      "epoch": 3.234782608695652,
+      "grad_norm": 1.4460102517208502,
+      "learning_rate": 3.0434915904486284e-06,
+      "loss": 0.358,
+      "step": 1860
+    },
+    {
+      "epoch": 3.252173913043478,
+      "grad_norm": 1.3048211800924494,
+      "learning_rate": 2.990691234537721e-06,
+      "loss": 0.3405,
+      "step": 1870
+    },
+    {
+      "epoch": 3.269565217391304,
+      "grad_norm": 1.3241574204938475,
+      "learning_rate": 2.938156766319156e-06,
+      "loss": 0.384,
+      "step": 1880
+    },
+    {
+      "epoch": 3.2869565217391306,
+      "grad_norm": 1.6150241974779345,
+      "learning_rate": 2.885895137570958e-06,
+      "loss": 0.3621,
+      "step": 1890
+    },
+    {
+      "epoch": 3.3043478260869565,
+      "grad_norm": 1.2579063705669289,
+      "learning_rate": 2.83391326396687e-06,
+      "loss": 0.3884,
+      "step": 1900
+    },
+    {
+      "epoch": 3.3217391304347825,
+      "grad_norm": 1.3801202825398762,
+      "learning_rate": 2.7822180241612077e-06,
+      "loss": 0.3798,
+      "step": 1910
+    },
+    {
+      "epoch": 3.3391304347826085,
+      "grad_norm": 1.396049263670027,
+      "learning_rate": 2.7308162588786303e-06,
+      "loss": 0.3605,
+      "step": 1920
+    },
+    {
+      "epoch": 3.356521739130435,
+      "grad_norm": 1.3748023066384665,
+      "learning_rate": 2.6797147700089167e-06,
+      "loss": 0.3579,
+      "step": 1930
+    },
+    {
+      "epoch": 3.373913043478261,
+      "grad_norm": 1.5175283161299356,
+      "learning_rate": 2.6289203197068834e-06,
+      "loss": 0.3784,
+      "step": 1940
+    },
+    {
+      "epoch": 3.391304347826087,
+      "grad_norm": 1.5851194282172685,
+      "learning_rate": 2.5784396294975677e-06,
+      "loss": 0.3606,
+      "step": 1950
+    },
+    {
+      "epoch": 3.408695652173913,
+      "grad_norm": 1.349779292865959,
+      "learning_rate": 2.528279379386783e-06,
+      "loss": 0.3694,
+      "step": 1960
+    },
+    {
+      "epoch": 3.426086956521739,
+      "grad_norm": 1.5339986272397241,
+      "learning_rate": 2.478446206977159e-06,
+      "loss": 0.3771,
+      "step": 1970
+    },
+    {
+      "epoch": 3.4434782608695653,
+      "grad_norm": 1.430106842741753,
+      "learning_rate": 2.4289467065898085e-06,
+      "loss": 0.369,
+      "step": 1980
+    },
+    {
+      "epoch": 3.4608695652173913,
+      "grad_norm": 1.457323252888418,
+      "learning_rate": 2.3797874283917127e-06,
+      "loss": 0.3403,
+      "step": 1990
+    },
+    {
+      "epoch": 3.4782608695652173,
+      "grad_norm": 1.4282143674816088,
+      "learning_rate": 2.3309748775289497e-06,
+      "loss": 0.3845,
+      "step": 2000
+    },
+    {
+      "epoch": 3.4956521739130437,
+      "grad_norm": 1.3173625440079775,
+      "learning_rate": 2.282515513265885e-06,
+      "loss": 0.3695,
+      "step": 2010
+    },
+    {
+      "epoch": 3.5130434782608697,
+      "grad_norm": 1.5391795918668048,
+      "learning_rate": 2.2344157481304267e-06,
+      "loss": 0.384,
+      "step": 2020
+    },
+    {
+      "epoch": 3.5304347826086957,
+      "grad_norm": 1.5462876476217569,
+      "learning_rate": 2.1866819470654727e-06,
+      "loss": 0.362,
+      "step": 2030
+    },
+    {
+      "epoch": 3.5478260869565217,
+      "grad_norm": 1.5819871351937547,
+      "learning_rate": 2.1393204265866467e-06,
+      "loss": 0.3788,
+      "step": 2040
+    },
+    {
+      "epoch": 3.5652173913043477,
+      "grad_norm": 1.6521343485618902,
+      "learning_rate": 2.09233745394645e-06,
+      "loss": 0.3785,
+      "step": 2050
+    },
+    {
+      "epoch": 3.5826086956521737,
+      "grad_norm": 1.5040618975765006,
+      "learning_rate": 2.0457392463049285e-06,
+      "loss": 0.359,
+      "step": 2060
+    },
+    {
+      "epoch": 3.6,
+      "grad_norm": 1.6074930743024214,
+      "learning_rate": 1.9995319699069664e-06,
+      "loss": 0.3652,
+      "step": 2070
+    },
+    {
+      "epoch": 3.617391304347826,
+      "grad_norm": 1.5631144357601194,
+      "learning_rate": 1.9537217392663218e-06,
+      "loss": 0.3626,
+      "step": 2080
+    },
+    {
+      "epoch": 3.634782608695652,
+      "grad_norm": 1.5044302087614625,
+      "learning_rate": 1.908314616356505e-06,
+      "loss": 0.388,
+      "step": 2090
+    },
+    {
+      "epoch": 3.6521739130434785,
+      "grad_norm": 1.3802858769348085,
+      "learning_rate": 1.8633166098086103e-06,
+      "loss": 0.3629,
+      "step": 2100
+    },
+    {
+      "epoch": 3.6695652173913045,
+      "grad_norm": 1.4222636474825245,
+      "learning_rate": 1.818733674116207e-06,
+      "loss": 0.3651,
+      "step": 2110
+    },
+    {
+      "epoch": 3.6869565217391305,
+      "grad_norm": 1.5026251469181622,
+      "learning_rate": 1.7745717088473895e-06,
+      "loss": 0.3772,
+      "step": 2120
+    },
+    {
+      "epoch": 3.7043478260869565,
+      "grad_norm": 1.4848721046911413,
+      "learning_rate": 1.7308365578641089e-06,
+      "loss": 0.3522,
+      "step": 2130
+    },
+    {
+      "epoch": 3.7217391304347824,
+      "grad_norm": 1.6975508268605253,
+      "learning_rate": 1.687534008548854e-06,
+      "loss": 0.3775,
+      "step": 2140
+    },
+    {
+      "epoch": 3.7391304347826084,
+      "grad_norm": 1.7810729043290516,
+      "learning_rate": 1.6446697910388294e-06,
+      "loss": 0.3647,
+      "step": 2150
+    },
+    {
+      "epoch": 3.756521739130435,
+      "grad_norm": 1.4737138340480456,
+      "learning_rate": 1.6022495774676916e-06,
+      "loss": 0.3723,
+      "step": 2160
+    },
+    {
+      "epoch": 3.773913043478261,
+      "grad_norm": 1.5559891939482013,
+      "learning_rate": 1.5602789812149727e-06,
+      "loss": 0.3503,
+      "step": 2170
+    },
+    {
+      "epoch": 3.791304347826087,
+      "grad_norm": 1.663018576508128,
+      "learning_rate": 1.5187635561632685e-06,
+      "loss": 0.3561,
+      "step": 2180
+    },
+    {
+      "epoch": 3.8086956521739133,
+      "grad_norm": 2.634784063202797,
+      "learning_rate": 1.477708795963308e-06,
+      "loss": 0.3406,
+      "step": 2190
+    },
+    {
+      "epoch": 3.8260869565217392,
+      "grad_norm": 1.5139630905983323,
+      "learning_rate": 1.4371201333069868e-06,
+      "loss": 0.3681,
+      "step": 2200
+    },
+    {
+      "epoch": 3.8434782608695652,
+      "grad_norm": 2.3176470378092424,
+      "learning_rate": 1.3970029392084771e-06,
+      "loss": 0.3541,
+      "step": 2210
+    },
+    {
+      "epoch": 3.860869565217391,
+      "grad_norm": 1.5959870558495743,
+      "learning_rate": 1.3573625222934829e-06,
+      "loss": 0.3675,
+      "step": 2220
+    },
+    {
+      "epoch": 3.878260869565217,
+      "grad_norm": 1.6337373914467612,
+      "learning_rate": 1.3182041280967656e-06,
+      "loss": 0.3514,
+      "step": 2230
+    },
+    {
+      "epoch": 3.8956521739130436,
+      "grad_norm": 1.4395314046176468,
+      "learning_rate": 1.2795329383680138e-06,
+      "loss": 0.3566,
+      "step": 2240
+    },
+    {
+      "epoch": 3.9130434782608696,
+      "grad_norm": 1.3199487970653823,
+      "learning_rate": 1.241354070386151e-06,
+      "loss": 0.3441,
+      "step": 2250
+    },
+    {
+      "epoch": 3.9304347826086956,
+      "grad_norm": 1.7513753951221092,
+      "learning_rate": 1.2036725762821783e-06,
+      "loss": 0.3628,
+      "step": 2260
+    },
+    {
+      "epoch": 3.9478260869565216,
+      "grad_norm": 1.5442153725537786,
+      "learning_rate": 1.1664934423706348e-06,
+      "loss": 0.3497,
+      "step": 2270
+    },
+    {
+      "epoch": 3.965217391304348,
+      "grad_norm": 1.663632062954429,
+      "learning_rate": 1.12982158848977e-06,
+      "loss": 0.3576,
+      "step": 2280
+    },
+    {
+      "epoch": 3.982608695652174,
+      "grad_norm": 1.8278674162694661,
+      "learning_rate": 1.0936618673505112e-06,
+      "loss": 0.3744,
+      "step": 2290
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.4687537010366873,
+      "learning_rate": 1.0580190638943138e-06,
+      "loss": 0.3549,
+      "step": 2300
+    },
+    {
+      "epoch": 4.017391304347826,
+      "grad_norm": 1.6716521197547283,
+      "learning_rate": 1.022897894659981e-06,
+      "loss": 0.3487,
+      "step": 2310
+    },
+    {
+      "epoch": 4.034782608695652,
+      "grad_norm": 1.4645341272970525,
+      "learning_rate": 9.883030071595335e-07,
+      "loss": 0.3158,
+      "step": 2320
+    },
+    {
+      "epoch": 4.052173913043478,
+      "grad_norm": 1.6557398280558575,
+      "learning_rate": 9.542389792632112e-07,
+      "loss": 0.331,
+      "step": 2330
+    },
+    {
+      "epoch": 4.069565217391304,
+      "grad_norm": 1.5960738974574165,
+      "learning_rate": 9.20710318593701e-07,
+      "loss": 0.3265,
+      "step": 2340
+    },
+    {
+      "epoch": 4.086956521739131,
+      "grad_norm": 1.4988502007449698,
+      "learning_rate": 8.877214619296421e-07,
+      "loss": 0.336,
+      "step": 2350
+    },
+    {
+      "epoch": 4.104347826086957,
+      "grad_norm": 1.4424789164433198,
+      "learning_rate": 8.552767746185215e-07,
+      "loss": 0.3347,
+      "step": 2360
+    },
+    {
+      "epoch": 4.121739130434783,
+      "grad_norm": 2.2719026550199475,
+      "learning_rate": 8.233805499990166e-07,
+      "loss": 0.3305,
+      "step": 2370
+    },
+    {
+      "epoch": 4.139130434782609,
+      "grad_norm": 1.4765476496178676,
+      "learning_rate": 7.920370088328672e-07,
+      "loss": 0.3484,
+      "step": 2380
+    },
+    {
+      "epoch": 4.156521739130435,
+      "grad_norm": 1.7030261009934748,
+      "learning_rate": 7.612502987463477e-07,
+      "loss": 0.3163,
+      "step": 2390
+    },
+    {
+      "epoch": 4.173913043478261,
+      "grad_norm": 1.4010667448317558,
+      "learning_rate": 7.310244936814232e-07,
+      "loss": 0.3222,
+      "step": 2400
+    },
+    {
+      "epoch": 4.191304347826087,
+      "grad_norm": 1.381057060755463,
+      "learning_rate": 7.013635933566515e-07,
+      "loss": 0.3175,
+      "step": 2410
+    },
+    {
+      "epoch": 4.208695652173913,
+      "grad_norm": 1.6489840526877189,
+      "learning_rate": 6.72271522737909e-07,
+      "loss": 0.3027,
+      "step": 2420
+    },
+    {
+      "epoch": 4.226086956521739,
+      "grad_norm": 1.4490054494783193,
+      "learning_rate": 6.437521315190087e-07,
+      "loss": 0.3529,
+      "step": 2430
+    },
+    {
+      "epoch": 4.243478260869566,
+      "grad_norm": 1.2850499028342035,
+      "learning_rate": 6.158091936122773e-07,
+      "loss": 0.3285,
+      "step": 2440
+    },
+    {
+      "epoch": 4.260869565217392,
+      "grad_norm": 1.2958305173488132,
+      "learning_rate": 5.884464066491613e-07,
+      "loss": 0.3128,
+      "step": 2450
+    },
+    {
+      "epoch": 4.278260869565218,
+      "grad_norm": 1.716194835927619,
+      "learning_rate": 5.616673914909282e-07,
+      "loss": 0.3377,
+      "step": 2460
+    },
+    {
+      "epoch": 4.2956521739130435,
+      "grad_norm": 1.5456459782523917,
+      "learning_rate": 5.354756917495224e-07,
+      "loss": 0.359,
+      "step": 2470
+    },
+    {
+      "epoch": 4.3130434782608695,
+      "grad_norm": 1.7049849318119408,
+      "learning_rate": 5.098747733186498e-07,
+      "loss": 0.3194,
+      "step": 2480
+    },
+    {
+      "epoch": 4.3304347826086955,
+      "grad_norm": 1.4376855175332917,
+      "learning_rate": 4.84868023915141e-07,
+      "loss": 0.3296,
+      "step": 2490
+    },
+    {
+      "epoch": 4.3478260869565215,
+      "grad_norm": 1.5575044356910057,
+      "learning_rate": 4.6045875263066474e-07,
+      "loss": 0.3347,
+      "step": 2500
+    },
+    {
+      "epoch": 4.3652173913043475,
+      "grad_norm": 1.6696231773916848,
+      "learning_rate": 4.366501894938363e-07,
+      "loss": 0.3509,
+      "step": 2510
+    },
+    {
+      "epoch": 4.3826086956521735,
+      "grad_norm": 1.4793874890497716,
+      "learning_rate": 4.1344548504280213e-07,
+      "loss": 0.316,
+      "step": 2520
+    },
+    {
+      "epoch": 4.4,
+      "grad_norm": 1.7504184745527158,
+      "learning_rate": 3.908477099083291e-07,
+      "loss": 0.3368,
+      "step": 2530
+    },
+    {
+      "epoch": 4.417391304347826,
+      "grad_norm": 1.5433746257360492,
+      "learning_rate": 3.6885985440747895e-07,
+      "loss": 0.3147,
+      "step": 2540
+    },
+    {
+      "epoch": 4.434782608695652,
+      "grad_norm": 1.6205451434238438,
+      "learning_rate": 3.474848281479032e-07,
+      "loss": 0.3198,
+      "step": 2550
+    },
+    {
+      "epoch": 4.452173913043478,
+      "grad_norm": 1.7748707264655563,
+      "learning_rate": 3.2672545964282263e-07,
+      "loss": 0.3337,
+      "step": 2560
+    },
+    {
+      "epoch": 4.469565217391304,
+      "grad_norm": 2.034670741566414,
+      "learning_rate": 3.0658449593673256e-07,
+      "loss": 0.2871,
+      "step": 2570
+    },
+    {
+      "epoch": 4.48695652173913,
+      "grad_norm": 1.9535751002955573,
+      "learning_rate": 2.8706460224189656e-07,
+      "loss": 0.3106,
+      "step": 2580
+    },
+    {
+      "epoch": 4.504347826086956,
+      "grad_norm": 1.4860584696429842,
+      "learning_rate": 2.6816836158566117e-07,
+      "loss": 0.3304,
+      "step": 2590
+    },
+    {
+      "epoch": 4.521739130434782,
+      "grad_norm": 1.5646385009356472,
+      "learning_rate": 2.498982744686501e-07,
+      "loss": 0.328,
+      "step": 2600
+    },
+    {
+      "epoch": 4.539130434782608,
+      "grad_norm": 1.6223923163321592,
+      "learning_rate": 2.3225675853387974e-07,
+      "loss": 0.3099,
+      "step": 2610
+    },
+    {
+      "epoch": 4.556521739130435,
+      "grad_norm": 1.4069397141672941,
+      "learning_rate": 2.1524614824683377e-07,
+      "loss": 0.3318,
+      "step": 2620
+    },
+    {
+      "epoch": 4.573913043478261,
+      "grad_norm": 1.5612850701477952,
+      "learning_rate": 1.98868694586552e-07,
+      "loss": 0.3376,
+      "step": 2630
+    },
+    {
+      "epoch": 4.591304347826087,
+      "grad_norm": 1.6586621274603865,
+      "learning_rate": 1.8312656474776093e-07,
+      "loss": 0.3034,
+      "step": 2640
+    },
+    {
+      "epoch": 4.608695652173913,
+      "grad_norm": 1.6277894314324821,
+      "learning_rate": 1.6802184185409355e-07,
+      "loss": 0.334,
+      "step": 2650
+    },
+    {
+      "epoch": 4.626086956521739,
+      "grad_norm": 1.5367253210879437,
+      "learning_rate": 1.5355652468243332e-07,
+      "loss": 0.3368,
+      "step": 2660
+    },
+    {
+      "epoch": 4.643478260869565,
+      "grad_norm": 1.42580964043298,
+      "learning_rate": 1.3973252739842236e-07,
+      "loss": 0.3107,
+      "step": 2670
+    },
+    {
+      "epoch": 4.660869565217391,
+      "grad_norm": 1.5836881272840495,
+      "learning_rate": 1.2655167930316236e-07,
+      "loss": 0.3337,
+      "step": 2680
+    },
+    {
+      "epoch": 4.678260869565217,
+      "grad_norm": 1.3933170878068741,
+      "learning_rate": 1.1401572459114441e-07,
+      "loss": 0.2988,
+      "step": 2690
+    },
+    {
+      "epoch": 4.695652173913043,
+      "grad_norm": 1.6629891174849243,
+      "learning_rate": 1.0212632211944906e-07,
+      "loss": 0.3038,
+      "step": 2700
+    },
+    {
+      "epoch": 4.71304347826087,
+      "grad_norm": 1.5180536343786482,
+      "learning_rate": 9.088504518822817e-08,
+      "loss": 0.2919,
+      "step": 2710
+    },
+    {
+      "epoch": 4.730434782608696,
+      "grad_norm": 1.4774316548557718,
+      "learning_rate": 8.029338133251518e-08,
+      "loss": 0.3253,
+      "step": 2720
+    },
+    {
+      "epoch": 4.747826086956522,
+      "grad_norm": 1.4916465022440153,
+      "learning_rate": 7.035273212538274e-08,
+      "loss": 0.3346,
+      "step": 2730
+    },
+    {
+      "epoch": 4.765217391304348,
+      "grad_norm": 1.6408617577143456,
+      "learning_rate": 6.10644129924759e-08,
+      "loss": 0.2921,
+      "step": 2740
+    },
+    {
+      "epoch": 4.782608695652174,
+      "grad_norm": 1.5935791526940084,
+      "learning_rate": 5.242965303794312e-08,
+      "loss": 0.3298,
+      "step": 2750
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 1.3615616479228225,
+      "learning_rate": 4.444959488179301e-08,
+      "loss": 0.3084,
+      "step": 2760
+    },
+    {
+      "epoch": 4.817391304347826,
+      "grad_norm": 1.3883499339712515,
+      "learning_rate": 3.7125294508693066e-08,
+      "loss": 0.3117,
+      "step": 2770
+    },
+    {
+      "epoch": 4.834782608695652,
+      "grad_norm": 1.483718903196712,
+      "learning_rate": 3.045772112823253e-08,
+      "loss": 0.3259,
+      "step": 2780
+    },
+    {
+      "epoch": 4.852173913043478,
+      "grad_norm": 1.5939760050066456,
+      "learning_rate": 2.4447757046670017e-08,
+      "loss": 0.3079,
+      "step": 2790
+    },
+    {
+      "epoch": 4.869565217391305,
+      "grad_norm": 1.5266765004604943,
+      "learning_rate": 1.9096197550179664e-08,
+      "loss": 0.3214,
+      "step": 2800
+    },
+    {
+      "epoch": 4.886956521739131,
+      "grad_norm": 1.7078593350970195,
+      "learning_rate": 1.4403750799613092e-08,
+      "loss": 0.3419,
+      "step": 2810
+    },
+    {
+      "epoch": 4.904347826086957,
+      "grad_norm": 1.595487144492393,
+      "learning_rate": 1.0371037736787714e-08,
+      "loss": 0.3224,
+      "step": 2820
+    },
+    {
+      "epoch": 4.921739130434783,
+      "grad_norm": 1.470874925227166,
+      "learning_rate": 6.998592002321336e-09,
+      "loss": 0.3303,
+      "step": 2830
+    },
+    {
+      "epoch": 4.939130434782609,
+      "grad_norm": 1.6419169053737916,
+      "learning_rate": 4.286859865014204e-09,
+      "loss": 0.3237,
+      "step": 2840
+    },
+    {
+      "epoch": 4.956521739130435,
+      "grad_norm": 1.7910736800765954,
+      "learning_rate": 2.236200162798463e-09,
+      "loss": 0.2943,
+      "step": 2850
+    },
+    {
+      "epoch": 4.973913043478261,
+      "grad_norm": 1.4483382289129645,
+      "learning_rate": 8.468842552505907e-10,
+      "loss": 0.3202,
+      "step": 2860
+    },
+    {
+      "epoch": 4.9913043478260875,
+      "grad_norm": 1.4884875727469478,
+      "learning_rate": 1.1909598768400366e-10,
+      "loss": 0.3389,
+      "step": 2870
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2875,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 10000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 33447808868352.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2875/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ce5cae7710446ef8c6957d71b42b5f7d1d0a240b0d25cb8116775791199f000
+size 7953

config.json

@@ -0,0 +1,28 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.52.3",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.52.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6299c5a2b8fb6b4c41b936e5153afbffc4dbeca795cb1e49eccdaaae8d95e486
+size 3087467144

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": 5.0,
+    "total_flos": 33447808868352.0,
+    "train_loss": 0.42618958398570184,
+    "train_runtime": 1921.5134,
+    "train_samples_per_second": 11.957,
+    "train_steps_per_second": 1.496
+}

trainer_log.jsonl

@@ -0,0 +1,288 @@
+{"current_steps": 10, "total_steps": 2875, "loss": 0.6248, "lr": 6.25e-07, "epoch": 0.017391304347826087, "percentage": 0.35, "elapsed_time": "0:00:11", "remaining_time": "0:53:17"}
+{"current_steps": 20, "total_steps": 2875, "loss": 0.6183, "lr": 1.3194444444444446e-06, "epoch": 0.034782608695652174, "percentage": 0.7, "elapsed_time": "0:00:18", "remaining_time": "0:43:49"}
+{"current_steps": 30, "total_steps": 2875, "loss": 0.6015, "lr": 2.0138888888888893e-06, "epoch": 0.05217391304347826, "percentage": 1.04, "elapsed_time": "0:00:25", "remaining_time": "0:40:33"}
+{"current_steps": 40, "total_steps": 2875, "loss": 0.5665, "lr": 2.7083333333333334e-06, "epoch": 0.06956521739130435, "percentage": 1.39, "elapsed_time": "0:00:32", "remaining_time": "0:38:04"}
+{"current_steps": 50, "total_steps": 2875, "loss": 0.54, "lr": 3.4027777777777783e-06, "epoch": 0.08695652173913043, "percentage": 1.74, "elapsed_time": "0:00:38", "remaining_time": "0:36:31"}
+{"current_steps": 60, "total_steps": 2875, "loss": 0.5399, "lr": 4.097222222222222e-06, "epoch": 0.10434782608695652, "percentage": 2.09, "elapsed_time": "0:00:45", "remaining_time": "0:35:23"}
+{"current_steps": 70, "total_steps": 2875, "loss": 0.5317, "lr": 4.791666666666668e-06, "epoch": 0.12173913043478261, "percentage": 2.43, "elapsed_time": "0:00:51", "remaining_time": "0:34:33"}
+{"current_steps": 80, "total_steps": 2875, "loss": 0.5183, "lr": 5.486111111111112e-06, "epoch": 0.1391304347826087, "percentage": 2.78, "elapsed_time": "0:00:58", "remaining_time": "0:33:54"}
+{"current_steps": 90, "total_steps": 2875, "loss": 0.5355, "lr": 6.180555555555557e-06, "epoch": 0.1565217391304348, "percentage": 3.13, "elapsed_time": "0:01:04", "remaining_time": "0:33:23"}
+{"current_steps": 100, "total_steps": 2875, "loss": 0.5215, "lr": 6.875e-06, "epoch": 0.17391304347826086, "percentage": 3.48, "elapsed_time": "0:01:11", "remaining_time": "0:32:59"}
+{"current_steps": 110, "total_steps": 2875, "loss": 0.5239, "lr": 7.569444444444445e-06, "epoch": 0.19130434782608696, "percentage": 3.83, "elapsed_time": "0:01:17", "remaining_time": "0:32:34"}
+{"current_steps": 120, "total_steps": 2875, "loss": 0.5288, "lr": 8.263888888888888e-06, "epoch": 0.20869565217391303, "percentage": 4.17, "elapsed_time": "0:01:24", "remaining_time": "0:32:14"}
+{"current_steps": 130, "total_steps": 2875, "loss": 0.5563, "lr": 8.958333333333334e-06, "epoch": 0.22608695652173913, "percentage": 4.52, "elapsed_time": "0:01:30", "remaining_time": "0:31:55"}
+{"current_steps": 140, "total_steps": 2875, "loss": 0.5047, "lr": 9.652777777777779e-06, "epoch": 0.24347826086956523, "percentage": 4.87, "elapsed_time": "0:01:37", "remaining_time": "0:31:40"}
+{"current_steps": 150, "total_steps": 2875, "loss": 0.5146, "lr": 9.999917294352674e-06, "epoch": 0.2608695652173913, "percentage": 5.22, "elapsed_time": "0:01:44", "remaining_time": "0:31:33"}
+{"current_steps": 160, "total_steps": 2875, "loss": 0.5342, "lr": 9.99925566559051e-06, "epoch": 0.2782608695652174, "percentage": 5.57, "elapsed_time": "0:01:50", "remaining_time": "0:31:22"}
+{"current_steps": 170, "total_steps": 2875, "loss": 0.5083, "lr": 9.997932495618156e-06, "epoch": 0.2956521739130435, "percentage": 5.91, "elapsed_time": "0:01:57", "remaining_time": "0:31:15"}
+{"current_steps": 180, "total_steps": 2875, "loss": 0.5063, "lr": 9.995947959527968e-06, "epoch": 0.3130434782608696, "percentage": 6.26, "elapsed_time": "0:02:04", "remaining_time": "0:31:09"}
+{"current_steps": 190, "total_steps": 2875, "loss": 0.5434, "lr": 9.993302319929523e-06, "epoch": 0.33043478260869563, "percentage": 6.61, "elapsed_time": "0:02:11", "remaining_time": "0:31:01"}
+{"current_steps": 200, "total_steps": 2875, "loss": 0.5189, "lr": 9.98999592691486e-06, "epoch": 0.34782608695652173, "percentage": 6.96, "elapsed_time": "0:02:18", "remaining_time": "0:30:53"}
+{"current_steps": 210, "total_steps": 2875, "loss": 0.5382, "lr": 9.986029218012164e-06, "epoch": 0.3652173913043478, "percentage": 7.3, "elapsed_time": "0:02:25", "remaining_time": "0:30:46"}
+{"current_steps": 220, "total_steps": 2875, "loss": 0.5276, "lr": 9.981402718127853e-06, "epoch": 0.3826086956521739, "percentage": 7.65, "elapsed_time": "0:02:32", "remaining_time": "0:30:39"}
+{"current_steps": 230, "total_steps": 2875, "loss": 0.5017, "lr": 9.976117039477133e-06, "epoch": 0.4, "percentage": 8.0, "elapsed_time": "0:02:39", "remaining_time": "0:30:31"}
+{"current_steps": 240, "total_steps": 2875, "loss": 0.4997, "lr": 9.970172881502982e-06, "epoch": 0.41739130434782606, "percentage": 8.35, "elapsed_time": "0:02:46", "remaining_time": "0:30:29"}
+{"current_steps": 250, "total_steps": 2875, "loss": 0.5582, "lr": 9.963571030783582e-06, "epoch": 0.43478260869565216, "percentage": 8.7, "elapsed_time": "0:02:53", "remaining_time": "0:30:22"}
+{"current_steps": 260, "total_steps": 2875, "loss": 0.5457, "lr": 9.956312360928253e-06, "epoch": 0.45217391304347826, "percentage": 9.04, "elapsed_time": "0:03:00", "remaining_time": "0:30:15"}
+{"current_steps": 270, "total_steps": 2875, "loss": 0.5552, "lr": 9.948397832461829e-06, "epoch": 0.46956521739130436, "percentage": 9.39, "elapsed_time": "0:03:07", "remaining_time": "0:30:07"}
+{"current_steps": 280, "total_steps": 2875, "loss": 0.5203, "lr": 9.93982849269757e-06, "epoch": 0.48695652173913045, "percentage": 9.74, "elapsed_time": "0:03:14", "remaining_time": "0:30:00"}
+{"current_steps": 290, "total_steps": 2875, "loss": 0.5409, "lr": 9.930605475598566e-06, "epoch": 0.5043478260869565, "percentage": 10.09, "elapsed_time": "0:03:21", "remaining_time": "0:29:53"}
+{"current_steps": 300, "total_steps": 2875, "loss": 0.5404, "lr": 9.92073000162768e-06, "epoch": 0.5217391304347826, "percentage": 10.43, "elapsed_time": "0:03:28", "remaining_time": "0:29:46"}
+{"current_steps": 310, "total_steps": 2875, "loss": 0.5188, "lr": 9.910203377586053e-06, "epoch": 0.5391304347826087, "percentage": 10.78, "elapsed_time": "0:03:35", "remaining_time": "0:29:39"}
+{"current_steps": 320, "total_steps": 2875, "loss": 0.4946, "lr": 9.899026996440173e-06, "epoch": 0.5565217391304348, "percentage": 11.13, "elapsed_time": "0:03:41", "remaining_time": "0:29:32"}
+{"current_steps": 330, "total_steps": 2875, "loss": 0.5298, "lr": 9.887202337137549e-06, "epoch": 0.5739130434782609, "percentage": 11.48, "elapsed_time": "0:03:50", "remaining_time": "0:29:34"}
+{"current_steps": 340, "total_steps": 2875, "loss": 0.5299, "lr": 9.874730964411001e-06, "epoch": 0.591304347826087, "percentage": 11.83, "elapsed_time": "0:03:59", "remaining_time": "0:29:43"}
+{"current_steps": 350, "total_steps": 2875, "loss": 0.5315, "lr": 9.861614528571607e-06, "epoch": 0.6086956521739131, "percentage": 12.17, "elapsed_time": "0:04:06", "remaining_time": "0:29:35"}
+{"current_steps": 360, "total_steps": 2875, "loss": 0.5447, "lr": 9.847854765290321e-06, "epoch": 0.6260869565217392, "percentage": 12.52, "elapsed_time": "0:04:13", "remaining_time": "0:29:27"}
+{"current_steps": 370, "total_steps": 2875, "loss": 0.5291, "lr": 9.83345349536829e-06, "epoch": 0.6434782608695652, "percentage": 12.87, "elapsed_time": "0:04:19", "remaining_time": "0:29:19"}
+{"current_steps": 380, "total_steps": 2875, "loss": 0.5107, "lr": 9.818412624495911e-06, "epoch": 0.6608695652173913, "percentage": 13.22, "elapsed_time": "0:04:26", "remaining_time": "0:29:11"}
+{"current_steps": 390, "total_steps": 2875, "loss": 0.5441, "lr": 9.802734143000668e-06, "epoch": 0.6782608695652174, "percentage": 13.57, "elapsed_time": "0:04:33", "remaining_time": "0:29:04"}
+{"current_steps": 400, "total_steps": 2875, "loss": 0.5403, "lr": 9.786420125583734e-06, "epoch": 0.6956521739130435, "percentage": 13.91, "elapsed_time": "0:04:40", "remaining_time": "0:28:56"}
+{"current_steps": 410, "total_steps": 2875, "loss": 0.5249, "lr": 9.769472731045451e-06, "epoch": 0.7130434782608696, "percentage": 14.26, "elapsed_time": "0:04:47", "remaining_time": "0:28:48"}
+{"current_steps": 420, "total_steps": 2875, "loss": 0.5399, "lr": 9.751894201999647e-06, "epoch": 0.7304347826086957, "percentage": 14.61, "elapsed_time": "0:04:54", "remaining_time": "0:28:40"}
+{"current_steps": 430, "total_steps": 2875, "loss": 0.5367, "lr": 9.733686864576883e-06, "epoch": 0.7478260869565218, "percentage": 14.96, "elapsed_time": "0:05:01", "remaining_time": "0:28:33"}
+{"current_steps": 440, "total_steps": 2875, "loss": 0.5251, "lr": 9.714853128116634e-06, "epoch": 0.7652173913043478, "percentage": 15.3, "elapsed_time": "0:05:08", "remaining_time": "0:28:25"}
+{"current_steps": 450, "total_steps": 2875, "loss": 0.555, "lr": 9.695395484848476e-06, "epoch": 0.782608695652174, "percentage": 15.65, "elapsed_time": "0:05:15", "remaining_time": "0:28:18"}
+{"current_steps": 460, "total_steps": 2875, "loss": 0.5379, "lr": 9.675316509562282e-06, "epoch": 0.8, "percentage": 16.0, "elapsed_time": "0:05:22", "remaining_time": "0:28:10"}
+{"current_steps": 470, "total_steps": 2875, "loss": 0.5115, "lr": 9.654618859267516e-06, "epoch": 0.8173913043478261, "percentage": 16.35, "elapsed_time": "0:05:28", "remaining_time": "0:28:03"}
+{"current_steps": 480, "total_steps": 2875, "loss": 0.5323, "lr": 9.633305272841632e-06, "epoch": 0.8347826086956521, "percentage": 16.7, "elapsed_time": "0:05:35", "remaining_time": "0:27:55"}
+{"current_steps": 490, "total_steps": 2875, "loss": 0.5379, "lr": 9.61137857066764e-06, "epoch": 0.8521739130434782, "percentage": 17.04, "elapsed_time": "0:05:42", "remaining_time": "0:27:48"}
+{"current_steps": 500, "total_steps": 2875, "loss": 0.5353, "lr": 9.5888416542609e-06, "epoch": 0.8695652173913043, "percentage": 17.39, "elapsed_time": "0:05:49", "remaining_time": "0:27:40"}
+{"current_steps": 510, "total_steps": 2875, "loss": 0.506, "lr": 9.565697505885165e-06, "epoch": 0.8869565217391304, "percentage": 17.74, "elapsed_time": "0:05:56", "remaining_time": "0:27:33"}
+{"current_steps": 520, "total_steps": 2875, "loss": 0.5291, "lr": 9.541949188157937e-06, "epoch": 0.9043478260869565, "percentage": 18.09, "elapsed_time": "0:06:03", "remaining_time": "0:27:26"}
+{"current_steps": 530, "total_steps": 2875, "loss": 0.51, "lr": 9.517599843645216e-06, "epoch": 0.9217391304347826, "percentage": 18.43, "elapsed_time": "0:06:12", "remaining_time": "0:27:27"}
+{"current_steps": 540, "total_steps": 2875, "loss": 0.5136, "lr": 9.492652694445629e-06, "epoch": 0.9391304347826087, "percentage": 18.78, "elapsed_time": "0:06:19", "remaining_time": "0:27:19"}
+{"current_steps": 550, "total_steps": 2875, "loss": 0.5402, "lr": 9.467111041764072e-06, "epoch": 0.9565217391304348, "percentage": 19.13, "elapsed_time": "0:06:26", "remaining_time": "0:27:12"}
+{"current_steps": 560, "total_steps": 2875, "loss": 0.5508, "lr": 9.44097826547486e-06, "epoch": 0.9739130434782609, "percentage": 19.48, "elapsed_time": "0:06:33", "remaining_time": "0:27:05"}
+{"current_steps": 570, "total_steps": 2875, "loss": 0.5516, "lr": 9.414257823674482e-06, "epoch": 0.991304347826087, "percentage": 19.83, "elapsed_time": "0:06:40", "remaining_time": "0:26:57"}
+{"current_steps": 580, "total_steps": 2875, "loss": 0.5011, "lr": 9.386953252223989e-06, "epoch": 1.008695652173913, "percentage": 20.17, "elapsed_time": "0:06:46", "remaining_time": "0:26:50"}
+{"current_steps": 590, "total_steps": 2875, "loss": 0.4851, "lr": 9.35906816428111e-06, "epoch": 1.0260869565217392, "percentage": 20.52, "elapsed_time": "0:06:53", "remaining_time": "0:26:42"}
+{"current_steps": 600, "total_steps": 2875, "loss": 0.4795, "lr": 9.330606249822125e-06, "epoch": 1.0434782608695652, "percentage": 20.87, "elapsed_time": "0:07:00", "remaining_time": "0:26:35"}
+{"current_steps": 610, "total_steps": 2875, "loss": 0.4955, "lr": 9.30157127515358e-06, "epoch": 1.0608695652173914, "percentage": 21.22, "elapsed_time": "0:07:07", "remaining_time": "0:26:28"}
+{"current_steps": 620, "total_steps": 2875, "loss": 0.4884, "lr": 9.271967082413899e-06, "epoch": 1.0782608695652174, "percentage": 21.57, "elapsed_time": "0:07:14", "remaining_time": "0:26:19"}
+{"current_steps": 630, "total_steps": 2875, "loss": 0.4655, "lr": 9.241797589064959e-06, "epoch": 1.0956521739130434, "percentage": 21.91, "elapsed_time": "0:07:20", "remaining_time": "0:26:10"}
+{"current_steps": 640, "total_steps": 2875, "loss": 0.4663, "lr": 9.211066787373702e-06, "epoch": 1.1130434782608696, "percentage": 22.26, "elapsed_time": "0:07:26", "remaining_time": "0:26:00"}
+{"current_steps": 650, "total_steps": 2875, "loss": 0.4725, "lr": 9.179778743883855e-06, "epoch": 1.1304347826086956, "percentage": 22.61, "elapsed_time": "0:07:33", "remaining_time": "0:25:51"}
+{"current_steps": 660, "total_steps": 2875, "loss": 0.4921, "lr": 9.147937598877797e-06, "epoch": 1.1478260869565218, "percentage": 22.96, "elapsed_time": "0:07:39", "remaining_time": "0:25:42"}
+{"current_steps": 670, "total_steps": 2875, "loss": 0.4649, "lr": 9.115547565828695e-06, "epoch": 1.1652173913043478, "percentage": 23.3, "elapsed_time": "0:07:46", "remaining_time": "0:25:34"}
+{"current_steps": 680, "total_steps": 2875, "loss": 0.5285, "lr": 9.082612930842942e-06, "epoch": 1.182608695652174, "percentage": 23.65, "elapsed_time": "0:07:52", "remaining_time": "0:25:25"}
+{"current_steps": 690, "total_steps": 2875, "loss": 0.4947, "lr": 9.049138052092982e-06, "epoch": 1.2, "percentage": 24.0, "elapsed_time": "0:07:59", "remaining_time": "0:25:17"}
+{"current_steps": 700, "total_steps": 2875, "loss": 0.4663, "lr": 9.015127359240603e-06, "epoch": 1.2173913043478262, "percentage": 24.35, "elapsed_time": "0:08:05", "remaining_time": "0:25:08"}
+{"current_steps": 710, "total_steps": 2875, "loss": 0.4828, "lr": 8.980585352850775e-06, "epoch": 1.2347826086956522, "percentage": 24.7, "elapsed_time": "0:08:12", "remaining_time": "0:25:00"}
+{"current_steps": 720, "total_steps": 2875, "loss": 0.4743, "lr": 8.94551660379609e-06, "epoch": 1.2521739130434781, "percentage": 25.04, "elapsed_time": "0:08:18", "remaining_time": "0:24:52"}
+{"current_steps": 730, "total_steps": 2875, "loss": 0.4996, "lr": 8.909925752651914e-06, "epoch": 1.2695652173913043, "percentage": 25.39, "elapsed_time": "0:08:25", "remaining_time": "0:24:43"}
+{"current_steps": 740, "total_steps": 2875, "loss": 0.4863, "lr": 8.873817509082305e-06, "epoch": 1.2869565217391306, "percentage": 25.74, "elapsed_time": "0:08:31", "remaining_time": "0:24:35"}
+{"current_steps": 750, "total_steps": 2875, "loss": 0.465, "lr": 8.837196651216802e-06, "epoch": 1.3043478260869565, "percentage": 26.09, "elapsed_time": "0:08:37", "remaining_time": "0:24:27"}
+{"current_steps": 760, "total_steps": 2875, "loss": 0.5031, "lr": 8.800068025018133e-06, "epoch": 1.3217391304347825, "percentage": 26.43, "elapsed_time": "0:08:44", "remaining_time": "0:24:19"}
+{"current_steps": 770, "total_steps": 2875, "loss": 0.4783, "lr": 8.762436543640965e-06, "epoch": 1.3391304347826087, "percentage": 26.78, "elapsed_time": "0:08:51", "remaining_time": "0:24:12"}
+{"current_steps": 780, "total_steps": 2875, "loss": 0.4883, "lr": 8.724307186781756e-06, "epoch": 1.3565217391304347, "percentage": 27.13, "elapsed_time": "0:08:57", "remaining_time": "0:24:03"}
+{"current_steps": 790, "total_steps": 2875, "loss": 0.4947, "lr": 8.685685000019803e-06, "epoch": 1.373913043478261, "percentage": 27.48, "elapsed_time": "0:09:04", "remaining_time": "0:23:57"}
+{"current_steps": 800, "total_steps": 2875, "loss": 0.4684, "lr": 8.646575094149568e-06, "epoch": 1.391304347826087, "percentage": 27.83, "elapsed_time": "0:09:12", "remaining_time": "0:23:53"}
+{"current_steps": 810, "total_steps": 2875, "loss": 0.4938, "lr": 8.606982644504378e-06, "epoch": 1.4086956521739131, "percentage": 28.17, "elapsed_time": "0:09:19", "remaining_time": "0:23:45"}
+{"current_steps": 820, "total_steps": 2875, "loss": 0.4723, "lr": 8.566912890271584e-06, "epoch": 1.4260869565217391, "percentage": 28.52, "elapsed_time": "0:09:25", "remaining_time": "0:23:37"}
+{"current_steps": 830, "total_steps": 2875, "loss": 0.5066, "lr": 8.526371133799277e-06, "epoch": 1.4434782608695653, "percentage": 28.87, "elapsed_time": "0:09:32", "remaining_time": "0:23:29"}
+{"current_steps": 840, "total_steps": 2875, "loss": 0.4809, "lr": 8.485362739894617e-06, "epoch": 1.4608695652173913, "percentage": 29.22, "elapsed_time": "0:09:38", "remaining_time": "0:23:21"}
+{"current_steps": 850, "total_steps": 2875, "loss": 0.4849, "lr": 8.443893135113956e-06, "epoch": 1.4782608695652173, "percentage": 29.57, "elapsed_time": "0:09:45", "remaining_time": "0:23:15"}
+{"current_steps": 860, "total_steps": 2875, "loss": 0.4731, "lr": 8.401967807044713e-06, "epoch": 1.4956521739130435, "percentage": 29.91, "elapsed_time": "0:09:52", "remaining_time": "0:23:07"}
+{"current_steps": 870, "total_steps": 2875, "loss": 0.4993, "lr": 8.359592303579241e-06, "epoch": 1.5130434782608697, "percentage": 30.26, "elapsed_time": "0:09:58", "remaining_time": "0:22:59"}
+{"current_steps": 880, "total_steps": 2875, "loss": 0.4831, "lr": 8.316772232180677e-06, "epoch": 1.5304347826086957, "percentage": 30.61, "elapsed_time": "0:10:05", "remaining_time": "0:22:51"}
+{"current_steps": 890, "total_steps": 2875, "loss": 0.4549, "lr": 8.273513259140911e-06, "epoch": 1.5478260869565217, "percentage": 30.96, "elapsed_time": "0:10:12", "remaining_time": "0:22:46"}
+{"current_steps": 900, "total_steps": 2875, "loss": 0.4889, "lr": 8.22982110883079e-06, "epoch": 1.5652173913043477, "percentage": 31.3, "elapsed_time": "0:10:19", "remaining_time": "0:22:39"}
+{"current_steps": 910, "total_steps": 2875, "loss": 0.507, "lr": 8.185701562942614e-06, "epoch": 1.5826086956521739, "percentage": 31.65, "elapsed_time": "0:10:25", "remaining_time": "0:22:31"}
+{"current_steps": 920, "total_steps": 2875, "loss": 0.4812, "lr": 8.141160459725063e-06, "epoch": 1.6, "percentage": 32.0, "elapsed_time": "0:10:32", "remaining_time": "0:22:23"}
+{"current_steps": 930, "total_steps": 2875, "loss": 0.4965, "lr": 8.096203693210626e-06, "epoch": 1.617391304347826, "percentage": 32.35, "elapsed_time": "0:10:40", "remaining_time": "0:22:18"}
+{"current_steps": 940, "total_steps": 2875, "loss": 0.4557, "lr": 8.050837212435662e-06, "epoch": 1.634782608695652, "percentage": 32.7, "elapsed_time": "0:10:46", "remaining_time": "0:22:11"}
+{"current_steps": 950, "total_steps": 2875, "loss": 0.4702, "lr": 8.00506702065318e-06, "epoch": 1.6521739130434783, "percentage": 33.04, "elapsed_time": "0:10:53", "remaining_time": "0:22:03"}
+{"current_steps": 960, "total_steps": 2875, "loss": 0.4695, "lr": 7.958899174538423e-06, "epoch": 1.6695652173913045, "percentage": 33.39, "elapsed_time": "0:10:59", "remaining_time": "0:21:55"}
+{"current_steps": 970, "total_steps": 2875, "loss": 0.4869, "lr": 7.912339783387429e-06, "epoch": 1.6869565217391305, "percentage": 33.74, "elapsed_time": "0:11:06", "remaining_time": "0:21:48"}
+{"current_steps": 980, "total_steps": 2875, "loss": 0.4556, "lr": 7.865395008308572e-06, "epoch": 1.7043478260869565, "percentage": 34.09, "elapsed_time": "0:11:12", "remaining_time": "0:21:40"}
+{"current_steps": 990, "total_steps": 2875, "loss": 0.5226, "lr": 7.818071061407295e-06, "epoch": 1.7217391304347827, "percentage": 34.43, "elapsed_time": "0:11:18", "remaining_time": "0:21:32"}
+{"current_steps": 1000, "total_steps": 2875, "loss": 0.4651, "lr": 7.770374204964062e-06, "epoch": 1.7391304347826086, "percentage": 34.78, "elapsed_time": "0:11:25", "remaining_time": "0:21:25"}
+{"current_steps": 1010, "total_steps": 2875, "loss": 0.498, "lr": 7.722310750605693e-06, "epoch": 1.7565217391304349, "percentage": 35.13, "elapsed_time": "0:11:32", "remaining_time": "0:21:17"}
+{"current_steps": 1020, "total_steps": 2875, "loss": 0.4832, "lr": 7.673887058470155e-06, "epoch": 1.7739130434782608, "percentage": 35.48, "elapsed_time": "0:11:38", "remaining_time": "0:21:10"}
+{"current_steps": 1030, "total_steps": 2875, "loss": 0.4716, "lr": 7.625109536364938e-06, "epoch": 1.7913043478260868, "percentage": 35.83, "elapsed_time": "0:11:45", "remaining_time": "0:21:02"}
+{"current_steps": 1040, "total_steps": 2875, "loss": 0.4696, "lr": 7.5759846389191234e-06, "epoch": 1.808695652173913, "percentage": 36.17, "elapsed_time": "0:11:51", "remaining_time": "0:20:55"}
+{"current_steps": 1050, "total_steps": 2875, "loss": 0.5235, "lr": 7.526518866729256e-06, "epoch": 1.8260869565217392, "percentage": 36.52, "elapsed_time": "0:11:58", "remaining_time": "0:20:48"}
+{"current_steps": 1060, "total_steps": 2875, "loss": 0.4936, "lr": 7.476718765499131e-06, "epoch": 1.8434782608695652, "percentage": 36.87, "elapsed_time": "0:12:04", "remaining_time": "0:20:40"}
+{"current_steps": 1070, "total_steps": 2875, "loss": 0.4954, "lr": 7.426590925173624e-06, "epoch": 1.8608695652173912, "percentage": 37.22, "elapsed_time": "0:12:11", "remaining_time": "0:20:33"}
+{"current_steps": 1080, "total_steps": 2875, "loss": 0.5226, "lr": 7.376141979066644e-06, "epoch": 1.8782608695652174, "percentage": 37.57, "elapsed_time": "0:12:17", "remaining_time": "0:20:25"}
+{"current_steps": 1090, "total_steps": 2875, "loss": 0.4997, "lr": 7.325378602983371e-06, "epoch": 1.8956521739130436, "percentage": 37.91, "elapsed_time": "0:12:24", "remaining_time": "0:20:18"}
+{"current_steps": 1100, "total_steps": 2875, "loss": 0.5069, "lr": 7.274307514336854e-06, "epoch": 1.9130434782608696, "percentage": 38.26, "elapsed_time": "0:12:30", "remaining_time": "0:20:11"}
+{"current_steps": 1110, "total_steps": 2875, "loss": 0.4516, "lr": 7.222935471259113e-06, "epoch": 1.9304347826086956, "percentage": 38.61, "elapsed_time": "0:12:37", "remaining_time": "0:20:03"}
+{"current_steps": 1120, "total_steps": 2875, "loss": 0.4771, "lr": 7.171269271706847e-06, "epoch": 1.9478260869565216, "percentage": 38.96, "elapsed_time": "0:12:43", "remaining_time": "0:19:56"}
+{"current_steps": 1130, "total_steps": 2875, "loss": 0.4628, "lr": 7.119315752561879e-06, "epoch": 1.9652173913043478, "percentage": 39.3, "elapsed_time": "0:12:50", "remaining_time": "0:19:49"}
+{"current_steps": 1140, "total_steps": 2875, "loss": 0.4961, "lr": 7.0670817887264375e-06, "epoch": 1.982608695652174, "percentage": 39.65, "elapsed_time": "0:12:56", "remaining_time": "0:19:42"}
+{"current_steps": 1150, "total_steps": 2875, "loss": 0.4854, "lr": 7.014574292213423e-06, "epoch": 2.0, "percentage": 40.0, "elapsed_time": "0:13:03", "remaining_time": "0:19:34"}
+{"current_steps": 1160, "total_steps": 2875, "loss": 0.4159, "lr": 6.9618002112317525e-06, "epoch": 2.017391304347826, "percentage": 40.35, "elapsed_time": "0:13:09", "remaining_time": "0:19:27"}
+{"current_steps": 1170, "total_steps": 2875, "loss": 0.4513, "lr": 6.908766529266915e-06, "epoch": 2.034782608695652, "percentage": 40.7, "elapsed_time": "0:13:16", "remaining_time": "0:19:20"}
+{"current_steps": 1180, "total_steps": 2875, "loss": 0.4147, "lr": 6.855480264156864e-06, "epoch": 2.0521739130434784, "percentage": 41.04, "elapsed_time": "0:13:22", "remaining_time": "0:19:12"}
+{"current_steps": 1190, "total_steps": 2875, "loss": 0.4279, "lr": 6.8019484671633586e-06, "epoch": 2.0695652173913044, "percentage": 41.39, "elapsed_time": "0:13:29", "remaining_time": "0:19:05"}
+{"current_steps": 1200, "total_steps": 2875, "loss": 0.4121, "lr": 6.748178222038891e-06, "epoch": 2.0869565217391304, "percentage": 41.74, "elapsed_time": "0:13:35", "remaining_time": "0:18:58"}
+{"current_steps": 1210, "total_steps": 2875, "loss": 0.4302, "lr": 6.694176644089297e-06, "epoch": 2.1043478260869564, "percentage": 42.09, "elapsed_time": "0:13:42", "remaining_time": "0:18:51"}
+{"current_steps": 1220, "total_steps": 2875, "loss": 0.4143, "lr": 6.639950879232218e-06, "epoch": 2.121739130434783, "percentage": 42.43, "elapsed_time": "0:13:48", "remaining_time": "0:18:43"}
+{"current_steps": 1230, "total_steps": 2875, "loss": 0.411, "lr": 6.585508103051478e-06, "epoch": 2.139130434782609, "percentage": 42.78, "elapsed_time": "0:13:55", "remaining_time": "0:18:36"}
+{"current_steps": 1240, "total_steps": 2875, "loss": 0.4247, "lr": 6.530855519847568e-06, "epoch": 2.1565217391304348, "percentage": 43.13, "elapsed_time": "0:14:01", "remaining_time": "0:18:29"}
+{"current_steps": 1250, "total_steps": 2875, "loss": 0.4, "lr": 6.4760003616843106e-06, "epoch": 2.1739130434782608, "percentage": 43.48, "elapsed_time": "0:14:07", "remaining_time": "0:18:22"}
+{"current_steps": 1260, "total_steps": 2875, "loss": 0.4016, "lr": 6.420949887431855e-06, "epoch": 2.1913043478260867, "percentage": 43.83, "elapsed_time": "0:14:14", "remaining_time": "0:18:15"}
+{"current_steps": 1270, "total_steps": 2875, "loss": 0.4408, "lr": 6.3657113818061225e-06, "epoch": 2.208695652173913, "percentage": 44.17, "elapsed_time": "0:14:20", "remaining_time": "0:18:07"}
+{"current_steps": 1280, "total_steps": 2875, "loss": 0.3999, "lr": 6.310292154404844e-06, "epoch": 2.226086956521739, "percentage": 44.52, "elapsed_time": "0:14:27", "remaining_time": "0:18:00"}
+{"current_steps": 1290, "total_steps": 2875, "loss": 0.4329, "lr": 6.254699538740292e-06, "epoch": 2.243478260869565, "percentage": 44.87, "elapsed_time": "0:14:33", "remaining_time": "0:17:53"}
+{"current_steps": 1300, "total_steps": 2875, "loss": 0.4282, "lr": 6.198940891268844e-06, "epoch": 2.260869565217391, "percentage": 45.22, "elapsed_time": "0:14:40", "remaining_time": "0:17:46"}
+{"current_steps": 1310, "total_steps": 2875, "loss": 0.4222, "lr": 6.14302359041753e-06, "epoch": 2.2782608695652176, "percentage": 45.57, "elapsed_time": "0:14:46", "remaining_time": "0:17:39"}
+{"current_steps": 1320, "total_steps": 2875, "loss": 0.4317, "lr": 6.086955035607655e-06, "epoch": 2.2956521739130435, "percentage": 45.91, "elapsed_time": "0:14:53", "remaining_time": "0:17:32"}
+{"current_steps": 1330, "total_steps": 2875, "loss": 0.4212, "lr": 6.030742646275647e-06, "epoch": 2.3130434782608695, "percentage": 46.26, "elapsed_time": "0:14:59", "remaining_time": "0:17:25"}
+{"current_steps": 1340, "total_steps": 2875, "loss": 0.4334, "lr": 5.9743938608912626e-06, "epoch": 2.3304347826086955, "percentage": 46.61, "elapsed_time": "0:15:06", "remaining_time": "0:17:18"}
+{"current_steps": 1350, "total_steps": 2875, "loss": 0.4153, "lr": 5.917916135973263e-06, "epoch": 2.3478260869565215, "percentage": 46.96, "elapsed_time": "0:15:12", "remaining_time": "0:17:10"}
+{"current_steps": 1360, "total_steps": 2875, "loss": 0.4103, "lr": 5.861316945102717e-06, "epoch": 2.365217391304348, "percentage": 47.3, "elapsed_time": "0:15:19", "remaining_time": "0:17:03"}
+{"current_steps": 1370, "total_steps": 2875, "loss": 0.4104, "lr": 5.804603777934032e-06, "epoch": 2.382608695652174, "percentage": 47.65, "elapsed_time": "0:15:27", "remaining_time": "0:16:58"}
+{"current_steps": 1380, "total_steps": 2875, "loss": 0.4245, "lr": 5.74778413920386e-06, "epoch": 2.4, "percentage": 48.0, "elapsed_time": "0:15:34", "remaining_time": "0:16:52"}
+{"current_steps": 1390, "total_steps": 2875, "loss": 0.4412, "lr": 5.690865547738021e-06, "epoch": 2.417391304347826, "percentage": 48.35, "elapsed_time": "0:15:40", "remaining_time": "0:16:45"}
+{"current_steps": 1400, "total_steps": 2875, "loss": 0.4281, "lr": 5.6338555354565445e-06, "epoch": 2.4347826086956523, "percentage": 48.7, "elapsed_time": "0:15:47", "remaining_time": "0:16:38"}
+{"current_steps": 1410, "total_steps": 2875, "loss": 0.4359, "lr": 5.576761646376987e-06, "epoch": 2.4521739130434783, "percentage": 49.04, "elapsed_time": "0:15:53", "remaining_time": "0:16:30"}
+{"current_steps": 1420, "total_steps": 2875, "loss": 0.4377, "lr": 5.519591435616153e-06, "epoch": 2.4695652173913043, "percentage": 49.39, "elapsed_time": "0:16:00", "remaining_time": "0:16:23"}
+{"current_steps": 1430, "total_steps": 2875, "loss": 0.4229, "lr": 5.462352468390333e-06, "epoch": 2.4869565217391303, "percentage": 49.74, "elapsed_time": "0:16:06", "remaining_time": "0:16:16"}
+{"current_steps": 1440, "total_steps": 2875, "loss": 0.4225, "lr": 5.405052319014223e-06, "epoch": 2.5043478260869563, "percentage": 50.09, "elapsed_time": "0:16:13", "remaining_time": "0:16:09"}
+{"current_steps": 1450, "total_steps": 2875, "loss": 0.4245, "lr": 5.347698569898624e-06, "epoch": 2.5217391304347827, "percentage": 50.43, "elapsed_time": "0:16:19", "remaining_time": "0:16:02"}
+{"current_steps": 1460, "total_steps": 2875, "loss": 0.3993, "lr": 5.290298810547083e-06, "epoch": 2.5391304347826087, "percentage": 50.78, "elapsed_time": "0:16:26", "remaining_time": "0:15:55"}
+{"current_steps": 1470, "total_steps": 2875, "loss": 0.4324, "lr": 5.232860636551583e-06, "epoch": 2.5565217391304347, "percentage": 51.13, "elapsed_time": "0:16:32", "remaining_time": "0:15:48"}
+{"current_steps": 1480, "total_steps": 2875, "loss": 0.4185, "lr": 5.175391648587443e-06, "epoch": 2.573913043478261, "percentage": 51.48, "elapsed_time": "0:16:39", "remaining_time": "0:15:41"}
+{"current_steps": 1490, "total_steps": 2875, "loss": 0.4309, "lr": 5.117899451407526e-06, "epoch": 2.591304347826087, "percentage": 51.83, "elapsed_time": "0:16:45", "remaining_time": "0:15:34"}
+{"current_steps": 1500, "total_steps": 2875, "loss": 0.4292, "lr": 5.060391652835925e-06, "epoch": 2.608695652173913, "percentage": 52.17, "elapsed_time": "0:16:52", "remaining_time": "0:15:27"}
+{"current_steps": 1510, "total_steps": 2875, "loss": 0.4268, "lr": 5.002875862761234e-06, "epoch": 2.626086956521739, "percentage": 52.52, "elapsed_time": "0:16:58", "remaining_time": "0:15:21"}
+{"current_steps": 1520, "total_steps": 2875, "loss": 0.4319, "lr": 4.9453596921295435e-06, "epoch": 2.643478260869565, "percentage": 52.87, "elapsed_time": "0:17:05", "remaining_time": "0:15:14"}
+{"current_steps": 1530, "total_steps": 2875, "loss": 0.4378, "lr": 4.8878507519373055e-06, "epoch": 2.660869565217391, "percentage": 53.22, "elapsed_time": "0:17:11", "remaining_time": "0:15:07"}
+{"current_steps": 1540, "total_steps": 2875, "loss": 0.4058, "lr": 4.830356652224181e-06, "epoch": 2.6782608695652175, "percentage": 53.57, "elapsed_time": "0:17:18", "remaining_time": "0:15:00"}
+{"current_steps": 1550, "total_steps": 2875, "loss": 0.4165, "lr": 4.77288500106602e-06, "epoch": 2.6956521739130435, "percentage": 53.91, "elapsed_time": "0:17:24", "remaining_time": "0:14:53"}
+{"current_steps": 1560, "total_steps": 2875, "loss": 0.4148, "lr": 4.715443403568103e-06, "epoch": 2.7130434782608694, "percentage": 54.26, "elapsed_time": "0:17:31", "remaining_time": "0:14:46"}
+{"current_steps": 1570, "total_steps": 2875, "loss": 0.4088, "lr": 4.65803946085877e-06, "epoch": 2.730434782608696, "percentage": 54.61, "elapsed_time": "0:17:37", "remaining_time": "0:14:39"}
+{"current_steps": 1580, "total_steps": 2875, "loss": 0.4359, "lr": 4.600680769083585e-06, "epoch": 2.747826086956522, "percentage": 54.96, "elapsed_time": "0:17:44", "remaining_time": "0:14:32"}
+{"current_steps": 1590, "total_steps": 2875, "loss": 0.4386, "lr": 4.543374918400142e-06, "epoch": 2.765217391304348, "percentage": 55.3, "elapsed_time": "0:17:50", "remaining_time": "0:14:25"}
+{"current_steps": 1600, "total_steps": 2875, "loss": 0.4517, "lr": 4.486129491973687e-06, "epoch": 2.782608695652174, "percentage": 55.65, "elapsed_time": "0:17:57", "remaining_time": "0:14:18"}
+{"current_steps": 1610, "total_steps": 2875, "loss": 0.4283, "lr": 4.4289520649736475e-06, "epoch": 2.8, "percentage": 56.0, "elapsed_time": "0:18:03", "remaining_time": "0:14:11"}
+{"current_steps": 1620, "total_steps": 2875, "loss": 0.421, "lr": 4.371850203571225e-06, "epoch": 2.8173913043478263, "percentage": 56.35, "elapsed_time": "0:18:10", "remaining_time": "0:14:04"}
+{"current_steps": 1630, "total_steps": 2875, "loss": 0.4188, "lr": 4.314831463938184e-06, "epoch": 2.8347826086956522, "percentage": 56.7, "elapsed_time": "0:18:16", "remaining_time": "0:13:57"}
+{"current_steps": 1640, "total_steps": 2875, "loss": 0.4094, "lr": 4.257903391246954e-06, "epoch": 2.8521739130434782, "percentage": 57.04, "elapsed_time": "0:18:23", "remaining_time": "0:13:50"}
+{"current_steps": 1650, "total_steps": 2875, "loss": 0.4411, "lr": 4.201073518672195e-06, "epoch": 2.869565217391304, "percentage": 57.39, "elapsed_time": "0:18:29", "remaining_time": "0:13:43"}
+{"current_steps": 1660, "total_steps": 2875, "loss": 0.4423, "lr": 4.144349366393949e-06, "epoch": 2.8869565217391306, "percentage": 57.74, "elapsed_time": "0:18:36", "remaining_time": "0:13:37"}
+{"current_steps": 1670, "total_steps": 2875, "loss": 0.4254, "lr": 4.08773844060251e-06, "epoch": 2.9043478260869566, "percentage": 58.09, "elapsed_time": "0:18:44", "remaining_time": "0:13:31"}
+{"current_steps": 1680, "total_steps": 2875, "loss": 0.4358, "lr": 4.031248232505139e-06, "epoch": 2.9217391304347826, "percentage": 58.43, "elapsed_time": "0:18:51", "remaining_time": "0:13:25"}
+{"current_steps": 1690, "total_steps": 2875, "loss": 0.4084, "lr": 3.97488621733478e-06, "epoch": 2.9391304347826086, "percentage": 58.78, "elapsed_time": "0:18:58", "remaining_time": "0:13:18"}
+{"current_steps": 1700, "total_steps": 2875, "loss": 0.4274, "lr": 3.918659853360864e-06, "epoch": 2.9565217391304346, "percentage": 59.13, "elapsed_time": "0:19:05", "remaining_time": "0:13:11"}
+{"current_steps": 1710, "total_steps": 2875, "loss": 0.4187, "lr": 3.862576580902383e-06, "epoch": 2.973913043478261, "percentage": 59.48, "elapsed_time": "0:19:12", "remaining_time": "0:13:04"}
+{"current_steps": 1720, "total_steps": 2875, "loss": 0.4303, "lr": 3.8066438213433234e-06, "epoch": 2.991304347826087, "percentage": 59.83, "elapsed_time": "0:19:19", "remaining_time": "0:12:58"}
+{"current_steps": 1730, "total_steps": 2875, "loss": 0.3752, "lr": 3.7508689761506055e-06, "epoch": 3.008695652173913, "percentage": 60.17, "elapsed_time": "0:19:25", "remaining_time": "0:12:51"}
+{"current_steps": 1740, "total_steps": 2875, "loss": 0.3434, "lr": 3.6952594258946693e-06, "epoch": 3.026086956521739, "percentage": 60.52, "elapsed_time": "0:19:31", "remaining_time": "0:12:44"}
+{"current_steps": 1750, "total_steps": 2875, "loss": 0.3607, "lr": 3.6398225292728185e-06, "epoch": 3.0434782608695654, "percentage": 60.87, "elapsed_time": "0:19:38", "remaining_time": "0:12:37"}
+{"current_steps": 1760, "total_steps": 2875, "loss": 0.3623, "lr": 3.584565622135453e-06, "epoch": 3.0608695652173914, "percentage": 61.22, "elapsed_time": "0:19:44", "remaining_time": "0:12:30"}
+{"current_steps": 1770, "total_steps": 2875, "loss": 0.3579, "lr": 3.5294960165153363e-06, "epoch": 3.0782608695652174, "percentage": 61.57, "elapsed_time": "0:19:51", "remaining_time": "0:12:23"}
+{"current_steps": 1780, "total_steps": 2875, "loss": 0.3973, "lr": 3.474620999660007e-06, "epoch": 3.0956521739130434, "percentage": 61.91, "elapsed_time": "0:19:58", "remaining_time": "0:12:17"}
+{"current_steps": 1790, "total_steps": 2875, "loss": 0.3579, "lr": 3.4199478330674745e-06, "epoch": 3.1130434782608694, "percentage": 62.26, "elapsed_time": "0:20:05", "remaining_time": "0:12:10"}
+{"current_steps": 1800, "total_steps": 2875, "loss": 0.3634, "lr": 3.365483751525317e-06, "epoch": 3.130434782608696, "percentage": 62.61, "elapsed_time": "0:20:11", "remaining_time": "0:12:03"}
+{"current_steps": 1810, "total_steps": 2875, "loss": 0.3427, "lr": 3.3112359621533193e-06, "epoch": 3.1478260869565218, "percentage": 62.96, "elapsed_time": "0:20:18", "remaining_time": "0:11:56"}
+{"current_steps": 1820, "total_steps": 2875, "loss": 0.3425, "lr": 3.257211643449768e-06, "epoch": 3.1652173913043478, "percentage": 63.3, "elapsed_time": "0:20:24", "remaining_time": "0:11:49"}
+{"current_steps": 1830, "total_steps": 2875, "loss": 0.3458, "lr": 3.203417944341536e-06, "epoch": 3.1826086956521737, "percentage": 63.65, "elapsed_time": "0:20:31", "remaining_time": "0:11:42"}
+{"current_steps": 1840, "total_steps": 2875, "loss": 0.3689, "lr": 3.149861983238082e-06, "epoch": 3.2, "percentage": 64.0, "elapsed_time": "0:20:37", "remaining_time": "0:11:36"}
+{"current_steps": 1850, "total_steps": 2875, "loss": 0.3521, "lr": 3.0965508470894812e-06, "epoch": 3.217391304347826, "percentage": 64.35, "elapsed_time": "0:20:44", "remaining_time": "0:11:29"}
+{"current_steps": 1860, "total_steps": 2875, "loss": 0.358, "lr": 3.0434915904486284e-06, "epoch": 3.234782608695652, "percentage": 64.7, "elapsed_time": "0:20:50", "remaining_time": "0:11:22"}
+{"current_steps": 1870, "total_steps": 2875, "loss": 0.3405, "lr": 2.990691234537721e-06, "epoch": 3.252173913043478, "percentage": 65.04, "elapsed_time": "0:20:56", "remaining_time": "0:11:15"}
+{"current_steps": 1880, "total_steps": 2875, "loss": 0.384, "lr": 2.938156766319156e-06, "epoch": 3.269565217391304, "percentage": 65.39, "elapsed_time": "0:21:03", "remaining_time": "0:11:08"}
+{"current_steps": 1890, "total_steps": 2875, "loss": 0.3621, "lr": 2.885895137570958e-06, "epoch": 3.2869565217391306, "percentage": 65.74, "elapsed_time": "0:21:09", "remaining_time": "0:11:01"}
+{"current_steps": 1900, "total_steps": 2875, "loss": 0.3884, "lr": 2.83391326396687e-06, "epoch": 3.3043478260869565, "percentage": 66.09, "elapsed_time": "0:21:16", "remaining_time": "0:10:55"}
+{"current_steps": 1910, "total_steps": 2875, "loss": 0.3798, "lr": 2.7822180241612077e-06, "epoch": 3.3217391304347825, "percentage": 66.43, "elapsed_time": "0:21:22", "remaining_time": "0:10:48"}
+{"current_steps": 1920, "total_steps": 2875, "loss": 0.3605, "lr": 2.7308162588786303e-06, "epoch": 3.3391304347826085, "percentage": 66.78, "elapsed_time": "0:21:29", "remaining_time": "0:10:41"}
+{"current_steps": 1930, "total_steps": 2875, "loss": 0.3579, "lr": 2.6797147700089167e-06, "epoch": 3.356521739130435, "percentage": 67.13, "elapsed_time": "0:21:35", "remaining_time": "0:10:34"}
+{"current_steps": 1940, "total_steps": 2875, "loss": 0.3784, "lr": 2.6289203197068834e-06, "epoch": 3.373913043478261, "percentage": 67.48, "elapsed_time": "0:21:42", "remaining_time": "0:10:27"}
+{"current_steps": 1950, "total_steps": 2875, "loss": 0.3606, "lr": 2.5784396294975677e-06, "epoch": 3.391304347826087, "percentage": 67.83, "elapsed_time": "0:21:48", "remaining_time": "0:10:20"}
+{"current_steps": 1960, "total_steps": 2875, "loss": 0.3694, "lr": 2.528279379386783e-06, "epoch": 3.408695652173913, "percentage": 68.17, "elapsed_time": "0:21:55", "remaining_time": "0:10:14"}
+{"current_steps": 1970, "total_steps": 2875, "loss": 0.3771, "lr": 2.478446206977159e-06, "epoch": 3.426086956521739, "percentage": 68.52, "elapsed_time": "0:22:02", "remaining_time": "0:10:07"}
+{"current_steps": 1980, "total_steps": 2875, "loss": 0.369, "lr": 2.4289467065898085e-06, "epoch": 3.4434782608695653, "percentage": 68.87, "elapsed_time": "0:22:09", "remaining_time": "0:10:01"}
+{"current_steps": 1990, "total_steps": 2875, "loss": 0.3403, "lr": 2.3797874283917127e-06, "epoch": 3.4608695652173913, "percentage": 69.22, "elapsed_time": "0:22:16", "remaining_time": "0:09:54"}
+{"current_steps": 2000, "total_steps": 2875, "loss": 0.3845, "lr": 2.3309748775289497e-06, "epoch": 3.4782608695652173, "percentage": 69.57, "elapsed_time": "0:22:22", "remaining_time": "0:09:47"}
+{"current_steps": 2010, "total_steps": 2875, "loss": 0.3695, "lr": 2.282515513265885e-06, "epoch": 3.4956521739130437, "percentage": 69.91, "elapsed_time": "0:22:29", "remaining_time": "0:09:40"}
+{"current_steps": 2020, "total_steps": 2875, "loss": 0.384, "lr": 2.2344157481304267e-06, "epoch": 3.5130434782608697, "percentage": 70.26, "elapsed_time": "0:22:35", "remaining_time": "0:09:33"}
+{"current_steps": 2030, "total_steps": 2875, "loss": 0.362, "lr": 2.1866819470654727e-06, "epoch": 3.5304347826086957, "percentage": 70.61, "elapsed_time": "0:22:42", "remaining_time": "0:09:27"}
+{"current_steps": 2040, "total_steps": 2875, "loss": 0.3788, "lr": 2.1393204265866467e-06, "epoch": 3.5478260869565217, "percentage": 70.96, "elapsed_time": "0:22:48", "remaining_time": "0:09:20"}
+{"current_steps": 2050, "total_steps": 2875, "loss": 0.3785, "lr": 2.09233745394645e-06, "epoch": 3.5652173913043477, "percentage": 71.3, "elapsed_time": "0:22:55", "remaining_time": "0:09:13"}
+{"current_steps": 2060, "total_steps": 2875, "loss": 0.359, "lr": 2.0457392463049285e-06, "epoch": 3.5826086956521737, "percentage": 71.65, "elapsed_time": "0:23:03", "remaining_time": "0:09:07"}
+{"current_steps": 2070, "total_steps": 2875, "loss": 0.3652, "lr": 1.9995319699069664e-06, "epoch": 3.6, "percentage": 72.0, "elapsed_time": "0:23:11", "remaining_time": "0:09:01"}
+{"current_steps": 2080, "total_steps": 2875, "loss": 0.3626, "lr": 1.9537217392663218e-06, "epoch": 3.617391304347826, "percentage": 72.35, "elapsed_time": "0:23:18", "remaining_time": "0:08:54"}
+{"current_steps": 2090, "total_steps": 2875, "loss": 0.388, "lr": 1.908314616356505e-06, "epoch": 3.634782608695652, "percentage": 72.7, "elapsed_time": "0:23:24", "remaining_time": "0:08:47"}
+{"current_steps": 2100, "total_steps": 2875, "loss": 0.3629, "lr": 1.8633166098086103e-06, "epoch": 3.6521739130434785, "percentage": 73.04, "elapsed_time": "0:23:31", "remaining_time": "0:08:40"}
+{"current_steps": 2110, "total_steps": 2875, "loss": 0.3651, "lr": 1.818733674116207e-06, "epoch": 3.6695652173913045, "percentage": 73.39, "elapsed_time": "0:23:37", "remaining_time": "0:08:34"}
+{"current_steps": 2120, "total_steps": 2875, "loss": 0.3772, "lr": 1.7745717088473895e-06, "epoch": 3.6869565217391305, "percentage": 73.74, "elapsed_time": "0:23:44", "remaining_time": "0:08:27"}
+{"current_steps": 2130, "total_steps": 2875, "loss": 0.3522, "lr": 1.7308365578641089e-06, "epoch": 3.7043478260869565, "percentage": 74.09, "elapsed_time": "0:23:50", "remaining_time": "0:08:20"}
+{"current_steps": 2140, "total_steps": 2875, "loss": 0.3775, "lr": 1.687534008548854e-06, "epoch": 3.7217391304347824, "percentage": 74.43, "elapsed_time": "0:23:57", "remaining_time": "0:08:13"}
+{"current_steps": 2150, "total_steps": 2875, "loss": 0.3647, "lr": 1.6446697910388294e-06, "epoch": 3.7391304347826084, "percentage": 74.78, "elapsed_time": "0:24:03", "remaining_time": "0:08:06"}
+{"current_steps": 2160, "total_steps": 2875, "loss": 0.3723, "lr": 1.6022495774676916e-06, "epoch": 3.756521739130435, "percentage": 75.13, "elapsed_time": "0:24:10", "remaining_time": "0:08:00"}
+{"current_steps": 2170, "total_steps": 2875, "loss": 0.3503, "lr": 1.5602789812149727e-06, "epoch": 3.773913043478261, "percentage": 75.48, "elapsed_time": "0:24:16", "remaining_time": "0:07:53"}
+{"current_steps": 2180, "total_steps": 2875, "loss": 0.3561, "lr": 1.5187635561632685e-06, "epoch": 3.791304347826087, "percentage": 75.83, "elapsed_time": "0:24:23", "remaining_time": "0:07:46"}
+{"current_steps": 2190, "total_steps": 2875, "loss": 0.3406, "lr": 1.477708795963308e-06, "epoch": 3.8086956521739133, "percentage": 76.17, "elapsed_time": "0:24:29", "remaining_time": "0:07:39"}
+{"current_steps": 2200, "total_steps": 2875, "loss": 0.3681, "lr": 1.4371201333069868e-06, "epoch": 3.8260869565217392, "percentage": 76.52, "elapsed_time": "0:24:36", "remaining_time": "0:07:32"}
+{"current_steps": 2210, "total_steps": 2875, "loss": 0.3541, "lr": 1.3970029392084771e-06, "epoch": 3.8434782608695652, "percentage": 76.87, "elapsed_time": "0:24:42", "remaining_time": "0:07:26"}
+{"current_steps": 2220, "total_steps": 2875, "loss": 0.3675, "lr": 1.3573625222934829e-06, "epoch": 3.860869565217391, "percentage": 77.22, "elapsed_time": "0:24:49", "remaining_time": "0:07:19"}
+{"current_steps": 2230, "total_steps": 2875, "loss": 0.3514, "lr": 1.3182041280967656e-06, "epoch": 3.878260869565217, "percentage": 77.57, "elapsed_time": "0:24:55", "remaining_time": "0:07:12"}
+{"current_steps": 2240, "total_steps": 2875, "loss": 0.3566, "lr": 1.2795329383680138e-06, "epoch": 3.8956521739130436, "percentage": 77.91, "elapsed_time": "0:25:02", "remaining_time": "0:07:05"}
+{"current_steps": 2250, "total_steps": 2875, "loss": 0.3441, "lr": 1.241354070386151e-06, "epoch": 3.9130434782608696, "percentage": 78.26, "elapsed_time": "0:25:08", "remaining_time": "0:06:59"}
+{"current_steps": 2260, "total_steps": 2875, "loss": 0.3628, "lr": 1.2036725762821783e-06, "epoch": 3.9304347826086956, "percentage": 78.61, "elapsed_time": "0:25:15", "remaining_time": "0:06:52"}
+{"current_steps": 2270, "total_steps": 2875, "loss": 0.3497, "lr": 1.1664934423706348e-06, "epoch": 3.9478260869565216, "percentage": 78.96, "elapsed_time": "0:25:22", "remaining_time": "0:06:45"}
+{"current_steps": 2280, "total_steps": 2875, "loss": 0.3576, "lr": 1.12982158848977e-06, "epoch": 3.965217391304348, "percentage": 79.3, "elapsed_time": "0:25:28", "remaining_time": "0:06:38"}
+{"current_steps": 2290, "total_steps": 2875, "loss": 0.3744, "lr": 1.0936618673505112e-06, "epoch": 3.982608695652174, "percentage": 79.65, "elapsed_time": "0:25:35", "remaining_time": "0:06:32"}
+{"current_steps": 2300, "total_steps": 2875, "loss": 0.3549, "lr": 1.0580190638943138e-06, "epoch": 4.0, "percentage": 80.0, "elapsed_time": "0:25:41", "remaining_time": "0:06:25"}
+{"current_steps": 2310, "total_steps": 2875, "loss": 0.3487, "lr": 1.022897894659981e-06, "epoch": 4.017391304347826, "percentage": 80.35, "elapsed_time": "0:25:48", "remaining_time": "0:06:18"}
+{"current_steps": 2320, "total_steps": 2875, "loss": 0.3158, "lr": 9.883030071595335e-07, "epoch": 4.034782608695652, "percentage": 80.7, "elapsed_time": "0:25:54", "remaining_time": "0:06:11"}
+{"current_steps": 2330, "total_steps": 2875, "loss": 0.331, "lr": 9.542389792632112e-07, "epoch": 4.052173913043478, "percentage": 81.04, "elapsed_time": "0:26:02", "remaining_time": "0:06:05"}
+{"current_steps": 2340, "total_steps": 2875, "loss": 0.3265, "lr": 9.20710318593701e-07, "epoch": 4.069565217391304, "percentage": 81.39, "elapsed_time": "0:26:09", "remaining_time": "0:05:58"}
+{"current_steps": 2350, "total_steps": 2875, "loss": 0.336, "lr": 8.877214619296421e-07, "epoch": 4.086956521739131, "percentage": 81.74, "elapsed_time": "0:26:15", "remaining_time": "0:05:51"}
+{"current_steps": 2360, "total_steps": 2875, "loss": 0.3347, "lr": 8.552767746185215e-07, "epoch": 4.104347826086957, "percentage": 82.09, "elapsed_time": "0:26:22", "remaining_time": "0:05:45"}
+{"current_steps": 2370, "total_steps": 2875, "loss": 0.3305, "lr": 8.233805499990166e-07, "epoch": 4.121739130434783, "percentage": 82.43, "elapsed_time": "0:26:28", "remaining_time": "0:05:38"}
+{"current_steps": 2380, "total_steps": 2875, "loss": 0.3484, "lr": 7.920370088328672e-07, "epoch": 4.139130434782609, "percentage": 82.78, "elapsed_time": "0:26:35", "remaining_time": "0:05:31"}
+{"current_steps": 2390, "total_steps": 2875, "loss": 0.3163, "lr": 7.612502987463477e-07, "epoch": 4.156521739130435, "percentage": 83.13, "elapsed_time": "0:26:41", "remaining_time": "0:05:25"}
+{"current_steps": 2400, "total_steps": 2875, "loss": 0.3222, "lr": 7.310244936814232e-07, "epoch": 4.173913043478261, "percentage": 83.48, "elapsed_time": "0:26:48", "remaining_time": "0:05:18"}
+{"current_steps": 2410, "total_steps": 2875, "loss": 0.3175, "lr": 7.013635933566515e-07, "epoch": 4.191304347826087, "percentage": 83.83, "elapsed_time": "0:26:55", "remaining_time": "0:05:11"}
+{"current_steps": 2420, "total_steps": 2875, "loss": 0.3027, "lr": 6.72271522737909e-07, "epoch": 4.208695652173913, "percentage": 84.17, "elapsed_time": "0:27:01", "remaining_time": "0:05:04"}
+{"current_steps": 2430, "total_steps": 2875, "loss": 0.3529, "lr": 6.437521315190087e-07, "epoch": 4.226086956521739, "percentage": 84.52, "elapsed_time": "0:27:08", "remaining_time": "0:04:58"}
+{"current_steps": 2440, "total_steps": 2875, "loss": 0.3285, "lr": 6.158091936122773e-07, "epoch": 4.243478260869566, "percentage": 84.87, "elapsed_time": "0:27:14", "remaining_time": "0:04:51"}
+{"current_steps": 2450, "total_steps": 2875, "loss": 0.3128, "lr": 5.884464066491613e-07, "epoch": 4.260869565217392, "percentage": 85.22, "elapsed_time": "0:27:21", "remaining_time": "0:04:44"}
+{"current_steps": 2460, "total_steps": 2875, "loss": 0.3377, "lr": 5.616673914909282e-07, "epoch": 4.278260869565218, "percentage": 85.57, "elapsed_time": "0:27:27", "remaining_time": "0:04:37"}
+{"current_steps": 2470, "total_steps": 2875, "loss": 0.359, "lr": 5.354756917495224e-07, "epoch": 4.2956521739130435, "percentage": 85.91, "elapsed_time": "0:27:34", "remaining_time": "0:04:31"}
+{"current_steps": 2480, "total_steps": 2875, "loss": 0.3194, "lr": 5.098747733186498e-07, "epoch": 4.3130434782608695, "percentage": 86.26, "elapsed_time": "0:27:40", "remaining_time": "0:04:24"}
+{"current_steps": 2490, "total_steps": 2875, "loss": 0.3296, "lr": 4.84868023915141e-07, "epoch": 4.3304347826086955, "percentage": 86.61, "elapsed_time": "0:27:47", "remaining_time": "0:04:17"}
+{"current_steps": 2500, "total_steps": 2875, "loss": 0.3347, "lr": 4.6045875263066474e-07, "epoch": 4.3478260869565215, "percentage": 86.96, "elapsed_time": "0:27:53", "remaining_time": "0:04:11"}
+{"current_steps": 2510, "total_steps": 2875, "loss": 0.3509, "lr": 4.366501894938363e-07, "epoch": 4.3652173913043475, "percentage": 87.3, "elapsed_time": "0:28:00", "remaining_time": "0:04:04"}
+{"current_steps": 2520, "total_steps": 2875, "loss": 0.316, "lr": 4.1344548504280213e-07, "epoch": 4.3826086956521735, "percentage": 87.65, "elapsed_time": "0:28:06", "remaining_time": "0:03:57"}
+{"current_steps": 2530, "total_steps": 2875, "loss": 0.3368, "lr": 3.908477099083291e-07, "epoch": 4.4, "percentage": 88.0, "elapsed_time": "0:28:13", "remaining_time": "0:03:50"}
+{"current_steps": 2540, "total_steps": 2875, "loss": 0.3147, "lr": 3.6885985440747895e-07, "epoch": 4.417391304347826, "percentage": 88.35, "elapsed_time": "0:28:19", "remaining_time": "0:03:44"}
+{"current_steps": 2550, "total_steps": 2875, "loss": 0.3198, "lr": 3.474848281479032e-07, "epoch": 4.434782608695652, "percentage": 88.7, "elapsed_time": "0:28:25", "remaining_time": "0:03:37"}
+{"current_steps": 2560, "total_steps": 2875, "loss": 0.3337, "lr": 3.2672545964282263e-07, "epoch": 4.452173913043478, "percentage": 89.04, "elapsed_time": "0:28:31", "remaining_time": "0:03:30"}
+{"current_steps": 2570, "total_steps": 2875, "loss": 0.2871, "lr": 3.0658449593673256e-07, "epoch": 4.469565217391304, "percentage": 89.39, "elapsed_time": "0:28:38", "remaining_time": "0:03:23"}
+{"current_steps": 2580, "total_steps": 2875, "loss": 0.3106, "lr": 2.8706460224189656e-07, "epoch": 4.48695652173913, "percentage": 89.74, "elapsed_time": "0:28:44", "remaining_time": "0:03:17"}
+{"current_steps": 2590, "total_steps": 2875, "loss": 0.3304, "lr": 2.6816836158566117e-07, "epoch": 4.504347826086956, "percentage": 90.09, "elapsed_time": "0:28:50", "remaining_time": "0:03:10"}
+{"current_steps": 2600, "total_steps": 2875, "loss": 0.328, "lr": 2.498982744686501e-07, "epoch": 4.521739130434782, "percentage": 90.43, "elapsed_time": "0:28:57", "remaining_time": "0:03:03"}
+{"current_steps": 2610, "total_steps": 2875, "loss": 0.3099, "lr": 2.3225675853387974e-07, "epoch": 4.539130434782608, "percentage": 90.78, "elapsed_time": "0:29:03", "remaining_time": "0:02:56"}
+{"current_steps": 2620, "total_steps": 2875, "loss": 0.3318, "lr": 2.1524614824683377e-07, "epoch": 4.556521739130435, "percentage": 91.13, "elapsed_time": "0:29:09", "remaining_time": "0:02:50"}
+{"current_steps": 2630, "total_steps": 2875, "loss": 0.3376, "lr": 1.98868694586552e-07, "epoch": 4.573913043478261, "percentage": 91.48, "elapsed_time": "0:29:15", "remaining_time": "0:02:43"}
+{"current_steps": 2640, "total_steps": 2875, "loss": 0.3034, "lr": 1.8312656474776093e-07, "epoch": 4.591304347826087, "percentage": 91.83, "elapsed_time": "0:29:22", "remaining_time": "0:02:36"}
+{"current_steps": 2650, "total_steps": 2875, "loss": 0.334, "lr": 1.6802184185409355e-07, "epoch": 4.608695652173913, "percentage": 92.17, "elapsed_time": "0:29:28", "remaining_time": "0:02:30"}
+{"current_steps": 2660, "total_steps": 2875, "loss": 0.3368, "lr": 1.5355652468243332e-07, "epoch": 4.626086956521739, "percentage": 92.52, "elapsed_time": "0:29:34", "remaining_time": "0:02:23"}
+{"current_steps": 2670, "total_steps": 2875, "loss": 0.3107, "lr": 1.3973252739842236e-07, "epoch": 4.643478260869565, "percentage": 92.87, "elapsed_time": "0:29:40", "remaining_time": "0:02:16"}
+{"current_steps": 2680, "total_steps": 2875, "loss": 0.3337, "lr": 1.2655167930316236e-07, "epoch": 4.660869565217391, "percentage": 93.22, "elapsed_time": "0:29:47", "remaining_time": "0:02:10"}
+{"current_steps": 2690, "total_steps": 2875, "loss": 0.2988, "lr": 1.1401572459114441e-07, "epoch": 4.678260869565217, "percentage": 93.57, "elapsed_time": "0:29:53", "remaining_time": "0:02:03"}
+{"current_steps": 2700, "total_steps": 2875, "loss": 0.3038, "lr": 1.0212632211944906e-07, "epoch": 4.695652173913043, "percentage": 93.91, "elapsed_time": "0:29:59", "remaining_time": "0:01:56"}
+{"current_steps": 2710, "total_steps": 2875, "loss": 0.2919, "lr": 9.088504518822817e-08, "epoch": 4.71304347826087, "percentage": 94.26, "elapsed_time": "0:30:05", "remaining_time": "0:01:49"}
+{"current_steps": 2720, "total_steps": 2875, "loss": 0.3253, "lr": 8.029338133251518e-08, "epoch": 4.730434782608696, "percentage": 94.61, "elapsed_time": "0:30:12", "remaining_time": "0:01:43"}
+{"current_steps": 2730, "total_steps": 2875, "loss": 0.3346, "lr": 7.035273212538274e-08, "epoch": 4.747826086956522, "percentage": 94.96, "elapsed_time": "0:30:18", "remaining_time": "0:01:36"}
+{"current_steps": 2740, "total_steps": 2875, "loss": 0.2921, "lr": 6.10644129924759e-08, "epoch": 4.765217391304348, "percentage": 95.3, "elapsed_time": "0:30:24", "remaining_time": "0:01:29"}
+{"current_steps": 2750, "total_steps": 2875, "loss": 0.3298, "lr": 5.242965303794312e-08, "epoch": 4.782608695652174, "percentage": 95.65, "elapsed_time": "0:30:31", "remaining_time": "0:01:23"}
+{"current_steps": 2760, "total_steps": 2875, "loss": 0.3084, "lr": 4.444959488179301e-08, "epoch": 4.8, "percentage": 96.0, "elapsed_time": "0:30:37", "remaining_time": "0:01:16"}
+{"current_steps": 2770, "total_steps": 2875, "loss": 0.3117, "lr": 3.7125294508693066e-08, "epoch": 4.817391304347826, "percentage": 96.35, "elapsed_time": "0:30:43", "remaining_time": "0:01:09"}
+{"current_steps": 2780, "total_steps": 2875, "loss": 0.3259, "lr": 3.045772112823253e-08, "epoch": 4.834782608695652, "percentage": 96.7, "elapsed_time": "0:30:50", "remaining_time": "0:01:03"}
+{"current_steps": 2790, "total_steps": 2875, "loss": 0.3079, "lr": 2.4447757046670017e-08, "epoch": 4.852173913043478, "percentage": 97.04, "elapsed_time": "0:30:58", "remaining_time": "0:00:56"}
+{"current_steps": 2800, "total_steps": 2875, "loss": 0.3214, "lr": 1.9096197550179664e-08, "epoch": 4.869565217391305, "percentage": 97.39, "elapsed_time": "0:31:04", "remaining_time": "0:00:49"}
+{"current_steps": 2810, "total_steps": 2875, "loss": 0.3419, "lr": 1.4403750799613092e-08, "epoch": 4.886956521739131, "percentage": 97.74, "elapsed_time": "0:31:11", "remaining_time": "0:00:43"}
+{"current_steps": 2820, "total_steps": 2875, "loss": 0.3224, "lr": 1.0371037736787714e-08, "epoch": 4.904347826086957, "percentage": 98.09, "elapsed_time": "0:31:18", "remaining_time": "0:00:36"}
+{"current_steps": 2830, "total_steps": 2875, "loss": 0.3303, "lr": 6.998592002321336e-09, "epoch": 4.921739130434783, "percentage": 98.43, "elapsed_time": "0:31:25", "remaining_time": "0:00:29"}
+{"current_steps": 2840, "total_steps": 2875, "loss": 0.3237, "lr": 4.286859865014204e-09, "epoch": 4.939130434782609, "percentage": 98.78, "elapsed_time": "0:31:31", "remaining_time": "0:00:23"}
+{"current_steps": 2850, "total_steps": 2875, "loss": 0.2943, "lr": 2.236200162798463e-09, "epoch": 4.956521739130435, "percentage": 99.13, "elapsed_time": "0:31:38", "remaining_time": "0:00:16"}
+{"current_steps": 2860, "total_steps": 2875, "loss": 0.3202, "lr": 8.468842552505907e-10, "epoch": 4.973913043478261, "percentage": 99.48, "elapsed_time": "0:31:44", "remaining_time": "0:00:09"}
+{"current_steps": 2870, "total_steps": 2875, "loss": 0.3389, "lr": 1.1909598768400366e-10, "epoch": 4.9913043478260875, "percentage": 99.83, "elapsed_time": "0:31:50", "remaining_time": "0:00:03"}
+{"current_steps": 2875, "total_steps": 2875, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "0:32:00", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,2052 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 2875,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.017391304347826087,
+      "grad_norm": 3.047056528200788,
+      "learning_rate": 6.25e-07,
+      "loss": 0.6248,
+      "step": 10
+    },
+    {
+      "epoch": 0.034782608695652174,
+      "grad_norm": 2.174143954299078,
+      "learning_rate": 1.3194444444444446e-06,
+      "loss": 0.6183,
+      "step": 20
+    },
+    {
+      "epoch": 0.05217391304347826,
+      "grad_norm": 1.864151160655997,
+      "learning_rate": 2.0138888888888893e-06,
+      "loss": 0.6015,
+      "step": 30
+    },
+    {
+      "epoch": 0.06956521739130435,
+      "grad_norm": 1.667310343818362,
+      "learning_rate": 2.7083333333333334e-06,
+      "loss": 0.5665,
+      "step": 40
+    },
+    {
+      "epoch": 0.08695652173913043,
+      "grad_norm": 2.455941646453557,
+      "learning_rate": 3.4027777777777783e-06,
+      "loss": 0.54,
+      "step": 50
+    },
+    {
+      "epoch": 0.10434782608695652,
+      "grad_norm": 1.9685053486872826,
+      "learning_rate": 4.097222222222222e-06,
+      "loss": 0.5399,
+      "step": 60
+    },
+    {
+      "epoch": 0.12173913043478261,
+      "grad_norm": 2.106997653377029,
+      "learning_rate": 4.791666666666668e-06,
+      "loss": 0.5317,
+      "step": 70
+    },
+    {
+      "epoch": 0.1391304347826087,
+      "grad_norm": 1.780173897349018,
+      "learning_rate": 5.486111111111112e-06,
+      "loss": 0.5183,
+      "step": 80
+    },
+    {
+      "epoch": 0.1565217391304348,
+      "grad_norm": 1.467101178589284,
+      "learning_rate": 6.180555555555557e-06,
+      "loss": 0.5355,
+      "step": 90
+    },
+    {
+      "epoch": 0.17391304347826086,
+      "grad_norm": 1.521288337808483,
+      "learning_rate": 6.875e-06,
+      "loss": 0.5215,
+      "step": 100
+    },
+    {
+      "epoch": 0.19130434782608696,
+      "grad_norm": 1.5769259173207397,
+      "learning_rate": 7.569444444444445e-06,
+      "loss": 0.5239,
+      "step": 110
+    },
+    {
+      "epoch": 0.20869565217391303,
+      "grad_norm": 1.2892203345431497,
+      "learning_rate": 8.263888888888888e-06,
+      "loss": 0.5288,
+      "step": 120
+    },
+    {
+      "epoch": 0.22608695652173913,
+      "grad_norm": 1.6642372146048716,
+      "learning_rate": 8.958333333333334e-06,
+      "loss": 0.5563,
+      "step": 130
+    },
+    {
+      "epoch": 0.24347826086956523,
+      "grad_norm": 1.5134595782170073,
+      "learning_rate": 9.652777777777779e-06,
+      "loss": 0.5047,
+      "step": 140
+    },
+    {
+      "epoch": 0.2608695652173913,
+      "grad_norm": 1.2945158944777713,
+      "learning_rate": 9.999917294352674e-06,
+      "loss": 0.5146,
+      "step": 150
+    },
+    {
+      "epoch": 0.2782608695652174,
+      "grad_norm": 1.7472246575863004,
+      "learning_rate": 9.99925566559051e-06,
+      "loss": 0.5342,
+      "step": 160
+    },
+    {
+      "epoch": 0.2956521739130435,
+      "grad_norm": 1.5193618638564104,
+      "learning_rate": 9.997932495618156e-06,
+      "loss": 0.5083,
+      "step": 170
+    },
+    {
+      "epoch": 0.3130434782608696,
+      "grad_norm": 1.554664459221582,
+      "learning_rate": 9.995947959527968e-06,
+      "loss": 0.5063,
+      "step": 180
+    },
+    {
+      "epoch": 0.33043478260869563,
+      "grad_norm": 2.1482072415475,
+      "learning_rate": 9.993302319929523e-06,
+      "loss": 0.5434,
+      "step": 190
+    },
+    {
+      "epoch": 0.34782608695652173,
+      "grad_norm": 1.4192382000638246,
+      "learning_rate": 9.98999592691486e-06,
+      "loss": 0.5189,
+      "step": 200
+    },
+    {
+      "epoch": 0.3652173913043478,
+      "grad_norm": 1.208599645402354,
+      "learning_rate": 9.986029218012164e-06,
+      "loss": 0.5382,
+      "step": 210
+    },
+    {
+      "epoch": 0.3826086956521739,
+      "grad_norm": 1.1836104891903814,
+      "learning_rate": 9.981402718127853e-06,
+      "loss": 0.5276,
+      "step": 220
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 1.5978655869742624,
+      "learning_rate": 9.976117039477133e-06,
+      "loss": 0.5017,
+      "step": 230
+    },
+    {
+      "epoch": 0.41739130434782606,
+      "grad_norm": 1.3359478565965037,
+      "learning_rate": 9.970172881502982e-06,
+      "loss": 0.4997,
+      "step": 240
+    },
+    {
+      "epoch": 0.43478260869565216,
+      "grad_norm": 1.3948206340232912,
+      "learning_rate": 9.963571030783582e-06,
+      "loss": 0.5582,
+      "step": 250
+    },
+    {
+      "epoch": 0.45217391304347826,
+      "grad_norm": 1.3011147589302579,
+      "learning_rate": 9.956312360928253e-06,
+      "loss": 0.5457,
+      "step": 260
+    },
+    {
+      "epoch": 0.46956521739130436,
+      "grad_norm": 1.080591596133825,
+      "learning_rate": 9.948397832461829e-06,
+      "loss": 0.5552,
+      "step": 270
+    },
+    {
+      "epoch": 0.48695652173913045,
+      "grad_norm": 1.2339442843093948,
+      "learning_rate": 9.93982849269757e-06,
+      "loss": 0.5203,
+      "step": 280
+    },
+    {
+      "epoch": 0.5043478260869565,
+      "grad_norm": 1.5073250471212296,
+      "learning_rate": 9.930605475598566e-06,
+      "loss": 0.5409,
+      "step": 290
+    },
+    {
+      "epoch": 0.5217391304347826,
+      "grad_norm": 1.2705670265704732,
+      "learning_rate": 9.92073000162768e-06,
+      "loss": 0.5404,
+      "step": 300
+    },
+    {
+      "epoch": 0.5391304347826087,
+      "grad_norm": 1.4491195774316425,
+      "learning_rate": 9.910203377586053e-06,
+      "loss": 0.5188,
+      "step": 310
+    },
+    {
+      "epoch": 0.5565217391304348,
+      "grad_norm": 1.4642066613650184,
+      "learning_rate": 9.899026996440173e-06,
+      "loss": 0.4946,
+      "step": 320
+    },
+    {
+      "epoch": 0.5739130434782609,
+      "grad_norm": 1.18880785751321,
+      "learning_rate": 9.887202337137549e-06,
+      "loss": 0.5298,
+      "step": 330
+    },
+    {
+      "epoch": 0.591304347826087,
+      "grad_norm": 1.2818806362850075,
+      "learning_rate": 9.874730964411001e-06,
+      "loss": 0.5299,
+      "step": 340
+    },
+    {
+      "epoch": 0.6086956521739131,
+      "grad_norm": 1.07998471259005,
+      "learning_rate": 9.861614528571607e-06,
+      "loss": 0.5315,
+      "step": 350
+    },
+    {
+      "epoch": 0.6260869565217392,
+      "grad_norm": 1.944925155516315,
+      "learning_rate": 9.847854765290321e-06,
+      "loss": 0.5447,
+      "step": 360
+    },
+    {
+      "epoch": 0.6434782608695652,
+      "grad_norm": 1.511844572828546,
+      "learning_rate": 9.83345349536829e-06,
+      "loss": 0.5291,
+      "step": 370
+    },
+    {
+      "epoch": 0.6608695652173913,
+      "grad_norm": 1.338025282900669,
+      "learning_rate": 9.818412624495911e-06,
+      "loss": 0.5107,
+      "step": 380
+    },
+    {
+      "epoch": 0.6782608695652174,
+      "grad_norm": 1.0620753499324938,
+      "learning_rate": 9.802734143000668e-06,
+      "loss": 0.5441,
+      "step": 390
+    },
+    {
+      "epoch": 0.6956521739130435,
+      "grad_norm": 1.3469601654140646,
+      "learning_rate": 9.786420125583734e-06,
+      "loss": 0.5403,
+      "step": 400
+    },
+    {
+      "epoch": 0.7130434782608696,
+      "grad_norm": 1.1557449160960933,
+      "learning_rate": 9.769472731045451e-06,
+      "loss": 0.5249,
+      "step": 410
+    },
+    {
+      "epoch": 0.7304347826086957,
+      "grad_norm": 1.275996117914776,
+      "learning_rate": 9.751894201999647e-06,
+      "loss": 0.5399,
+      "step": 420
+    },
+    {
+      "epoch": 0.7478260869565218,
+      "grad_norm": 1.2172722683153971,
+      "learning_rate": 9.733686864576883e-06,
+      "loss": 0.5367,
+      "step": 430
+    },
+    {
+      "epoch": 0.7652173913043478,
+      "grad_norm": 1.3421828009403691,
+      "learning_rate": 9.714853128116634e-06,
+      "loss": 0.5251,
+      "step": 440
+    },
+    {
+      "epoch": 0.782608695652174,
+      "grad_norm": 1.0911936537573848,
+      "learning_rate": 9.695395484848476e-06,
+      "loss": 0.555,
+      "step": 450
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 1.3133546561526954,
+      "learning_rate": 9.675316509562282e-06,
+      "loss": 0.5379,
+      "step": 460
+    },
+    {
+      "epoch": 0.8173913043478261,
+      "grad_norm": 1.0450523994543341,
+      "learning_rate": 9.654618859267516e-06,
+      "loss": 0.5115,
+      "step": 470
+    },
+    {
+      "epoch": 0.8347826086956521,
+      "grad_norm": 1.0976625003665517,
+      "learning_rate": 9.633305272841632e-06,
+      "loss": 0.5323,
+      "step": 480
+    },
+    {
+      "epoch": 0.8521739130434782,
+      "grad_norm": 1.1811352794597962,
+      "learning_rate": 9.61137857066764e-06,
+      "loss": 0.5379,
+      "step": 490
+    },
+    {
+      "epoch": 0.8695652173913043,
+      "grad_norm": 1.3489402359600888,
+      "learning_rate": 9.5888416542609e-06,
+      "loss": 0.5353,
+      "step": 500
+    },
+    {
+      "epoch": 0.8869565217391304,
+      "grad_norm": 1.2068307201497395,
+      "learning_rate": 9.565697505885165e-06,
+      "loss": 0.506,
+      "step": 510
+    },
+    {
+      "epoch": 0.9043478260869565,
+      "grad_norm": 1.2234807296718222,
+      "learning_rate": 9.541949188157937e-06,
+      "loss": 0.5291,
+      "step": 520
+    },
+    {
+      "epoch": 0.9217391304347826,
+      "grad_norm": 1.2256177402823996,
+      "learning_rate": 9.517599843645216e-06,
+      "loss": 0.51,
+      "step": 530
+    },
+    {
+      "epoch": 0.9391304347826087,
+      "grad_norm": 1.2281368130382897,
+      "learning_rate": 9.492652694445629e-06,
+      "loss": 0.5136,
+      "step": 540
+    },
+    {
+      "epoch": 0.9565217391304348,
+      "grad_norm": 1.1730852438772608,
+      "learning_rate": 9.467111041764072e-06,
+      "loss": 0.5402,
+      "step": 550
+    },
+    {
+      "epoch": 0.9739130434782609,
+      "grad_norm": 1.0983121942778715,
+      "learning_rate": 9.44097826547486e-06,
+      "loss": 0.5508,
+      "step": 560
+    },
+    {
+      "epoch": 0.991304347826087,
+      "grad_norm": 1.1697621737952535,
+      "learning_rate": 9.414257823674482e-06,
+      "loss": 0.5516,
+      "step": 570
+    },
+    {
+      "epoch": 1.008695652173913,
+      "grad_norm": 1.2095450172211872,
+      "learning_rate": 9.386953252223989e-06,
+      "loss": 0.5011,
+      "step": 580
+    },
+    {
+      "epoch": 1.0260869565217392,
+      "grad_norm": 1.1615174794799035,
+      "learning_rate": 9.35906816428111e-06,
+      "loss": 0.4851,
+      "step": 590
+    },
+    {
+      "epoch": 1.0434782608695652,
+      "grad_norm": 1.0792347526096802,
+      "learning_rate": 9.330606249822125e-06,
+      "loss": 0.4795,
+      "step": 600
+    },
+    {
+      "epoch": 1.0608695652173914,
+      "grad_norm": 1.097575063279078,
+      "learning_rate": 9.30157127515358e-06,
+      "loss": 0.4955,
+      "step": 610
+    },
+    {
+      "epoch": 1.0782608695652174,
+      "grad_norm": 1.1031575053788267,
+      "learning_rate": 9.271967082413899e-06,
+      "loss": 0.4884,
+      "step": 620
+    },
+    {
+      "epoch": 1.0956521739130434,
+      "grad_norm": 1.2767392201135948,
+      "learning_rate": 9.241797589064959e-06,
+      "loss": 0.4655,
+      "step": 630
+    },
+    {
+      "epoch": 1.1130434782608696,
+      "grad_norm": 1.1033082483572434,
+      "learning_rate": 9.211066787373702e-06,
+      "loss": 0.4663,
+      "step": 640
+    },
+    {
+      "epoch": 1.1304347826086956,
+      "grad_norm": 1.576080502703696,
+      "learning_rate": 9.179778743883855e-06,
+      "loss": 0.4725,
+      "step": 650
+    },
+    {
+      "epoch": 1.1478260869565218,
+      "grad_norm": 1.175175499900073,
+      "learning_rate": 9.147937598877797e-06,
+      "loss": 0.4921,
+      "step": 660
+    },
+    {
+      "epoch": 1.1652173913043478,
+      "grad_norm": 1.1370691886093878,
+      "learning_rate": 9.115547565828695e-06,
+      "loss": 0.4649,
+      "step": 670
+    },
+    {
+      "epoch": 1.182608695652174,
+      "grad_norm": 1.2601647572096586,
+      "learning_rate": 9.082612930842942e-06,
+      "loss": 0.5285,
+      "step": 680
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 1.2187851162701455,
+      "learning_rate": 9.049138052092982e-06,
+      "loss": 0.4947,
+      "step": 690
+    },
+    {
+      "epoch": 1.2173913043478262,
+      "grad_norm": 1.1224072986238753,
+      "learning_rate": 9.015127359240603e-06,
+      "loss": 0.4663,
+      "step": 700
+    },
+    {
+      "epoch": 1.2347826086956522,
+      "grad_norm": 1.2366635360634342,
+      "learning_rate": 8.980585352850775e-06,
+      "loss": 0.4828,
+      "step": 710
+    },
+    {
+      "epoch": 1.2521739130434781,
+      "grad_norm": 1.4981539175915906,
+      "learning_rate": 8.94551660379609e-06,
+      "loss": 0.4743,
+      "step": 720
+    },
+    {
+      "epoch": 1.2695652173913043,
+      "grad_norm": 1.0729421363170109,
+      "learning_rate": 8.909925752651914e-06,
+      "loss": 0.4996,
+      "step": 730
+    },
+    {
+      "epoch": 1.2869565217391306,
+      "grad_norm": 1.1955218613972736,
+      "learning_rate": 8.873817509082305e-06,
+      "loss": 0.4863,
+      "step": 740
+    },
+    {
+      "epoch": 1.3043478260869565,
+      "grad_norm": 1.126282527653828,
+      "learning_rate": 8.837196651216802e-06,
+      "loss": 0.465,
+      "step": 750
+    },
+    {
+      "epoch": 1.3217391304347825,
+      "grad_norm": 1.253199532462182,
+      "learning_rate": 8.800068025018133e-06,
+      "loss": 0.5031,
+      "step": 760
+    },
+    {
+      "epoch": 1.3391304347826087,
+      "grad_norm": 1.1627263868301658,
+      "learning_rate": 8.762436543640965e-06,
+      "loss": 0.4783,
+      "step": 770
+    },
+    {
+      "epoch": 1.3565217391304347,
+      "grad_norm": 1.2722684925328063,
+      "learning_rate": 8.724307186781756e-06,
+      "loss": 0.4883,
+      "step": 780
+    },
+    {
+      "epoch": 1.373913043478261,
+      "grad_norm": 1.1113001887408591,
+      "learning_rate": 8.685685000019803e-06,
+      "loss": 0.4947,
+      "step": 790
+    },
+    {
+      "epoch": 1.391304347826087,
+      "grad_norm": 1.0089575387474536,
+      "learning_rate": 8.646575094149568e-06,
+      "loss": 0.4684,
+      "step": 800
+    },
+    {
+      "epoch": 1.4086956521739131,
+      "grad_norm": 1.139961867951673,
+      "learning_rate": 8.606982644504378e-06,
+      "loss": 0.4938,
+      "step": 810
+    },
+    {
+      "epoch": 1.4260869565217391,
+      "grad_norm": 1.099302351381361,
+      "learning_rate": 8.566912890271584e-06,
+      "loss": 0.4723,
+      "step": 820
+    },
+    {
+      "epoch": 1.4434782608695653,
+      "grad_norm": 1.097933896784172,
+      "learning_rate": 8.526371133799277e-06,
+      "loss": 0.5066,
+      "step": 830
+    },
+    {
+      "epoch": 1.4608695652173913,
+      "grad_norm": 1.2653660292097024,
+      "learning_rate": 8.485362739894617e-06,
+      "loss": 0.4809,
+      "step": 840
+    },
+    {
+      "epoch": 1.4782608695652173,
+      "grad_norm": 1.031416554665846,
+      "learning_rate": 8.443893135113956e-06,
+      "loss": 0.4849,
+      "step": 850
+    },
+    {
+      "epoch": 1.4956521739130435,
+      "grad_norm": 1.2133975894170272,
+      "learning_rate": 8.401967807044713e-06,
+      "loss": 0.4731,
+      "step": 860
+    },
+    {
+      "epoch": 1.5130434782608697,
+      "grad_norm": 1.3527575598259336,
+      "learning_rate": 8.359592303579241e-06,
+      "loss": 0.4993,
+      "step": 870
+    },
+    {
+      "epoch": 1.5304347826086957,
+      "grad_norm": 1.5538636433489414,
+      "learning_rate": 8.316772232180677e-06,
+      "loss": 0.4831,
+      "step": 880
+    },
+    {
+      "epoch": 1.5478260869565217,
+      "grad_norm": 1.0668506396404902,
+      "learning_rate": 8.273513259140911e-06,
+      "loss": 0.4549,
+      "step": 890
+    },
+    {
+      "epoch": 1.5652173913043477,
+      "grad_norm": 1.3545786110485585,
+      "learning_rate": 8.22982110883079e-06,
+      "loss": 0.4889,
+      "step": 900
+    },
+    {
+      "epoch": 1.5826086956521739,
+      "grad_norm": 1.343041115784667,
+      "learning_rate": 8.185701562942614e-06,
+      "loss": 0.507,
+      "step": 910
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 1.2218861872968096,
+      "learning_rate": 8.141160459725063e-06,
+      "loss": 0.4812,
+      "step": 920
+    },
+    {
+      "epoch": 1.617391304347826,
+      "grad_norm": 1.2015960018363139,
+      "learning_rate": 8.096203693210626e-06,
+      "loss": 0.4965,
+      "step": 930
+    },
+    {
+      "epoch": 1.634782608695652,
+      "grad_norm": 1.096282312846291,
+      "learning_rate": 8.050837212435662e-06,
+      "loss": 0.4557,
+      "step": 940
+    },
+    {
+      "epoch": 1.6521739130434783,
+      "grad_norm": 1.0909506176288037,
+      "learning_rate": 8.00506702065318e-06,
+      "loss": 0.4702,
+      "step": 950
+    },
+    {
+      "epoch": 1.6695652173913045,
+      "grad_norm": 1.5020729724704236,
+      "learning_rate": 7.958899174538423e-06,
+      "loss": 0.4695,
+      "step": 960
+    },
+    {
+      "epoch": 1.6869565217391305,
+      "grad_norm": 1.1184823557537737,
+      "learning_rate": 7.912339783387429e-06,
+      "loss": 0.4869,
+      "step": 970
+    },
+    {
+      "epoch": 1.7043478260869565,
+      "grad_norm": 1.5926675735162503,
+      "learning_rate": 7.865395008308572e-06,
+      "loss": 0.4556,
+      "step": 980
+    },
+    {
+      "epoch": 1.7217391304347827,
+      "grad_norm": 1.357255668450024,
+      "learning_rate": 7.818071061407295e-06,
+      "loss": 0.5226,
+      "step": 990
+    },
+    {
+      "epoch": 1.7391304347826086,
+      "grad_norm": 1.1738545925108272,
+      "learning_rate": 7.770374204964062e-06,
+      "loss": 0.4651,
+      "step": 1000
+    },
+    {
+      "epoch": 1.7565217391304349,
+      "grad_norm": 1.1820022034633777,
+      "learning_rate": 7.722310750605693e-06,
+      "loss": 0.498,
+      "step": 1010
+    },
+    {
+      "epoch": 1.7739130434782608,
+      "grad_norm": 1.3439499575644294,
+      "learning_rate": 7.673887058470155e-06,
+      "loss": 0.4832,
+      "step": 1020
+    },
+    {
+      "epoch": 1.7913043478260868,
+      "grad_norm": 1.5304830271309717,
+      "learning_rate": 7.625109536364938e-06,
+      "loss": 0.4716,
+      "step": 1030
+    },
+    {
+      "epoch": 1.808695652173913,
+      "grad_norm": 1.2364034202681973,
+      "learning_rate": 7.5759846389191234e-06,
+      "loss": 0.4696,
+      "step": 1040
+    },
+    {
+      "epoch": 1.8260869565217392,
+      "grad_norm": 1.0229154001597585,
+      "learning_rate": 7.526518866729256e-06,
+      "loss": 0.5235,
+      "step": 1050
+    },
+    {
+      "epoch": 1.8434782608695652,
+      "grad_norm": 1.7878383661619177,
+      "learning_rate": 7.476718765499131e-06,
+      "loss": 0.4936,
+      "step": 1060
+    },
+    {
+      "epoch": 1.8608695652173912,
+      "grad_norm": 1.1220136346350835,
+      "learning_rate": 7.426590925173624e-06,
+      "loss": 0.4954,
+      "step": 1070
+    },
+    {
+      "epoch": 1.8782608695652174,
+      "grad_norm": 1.0276108610586765,
+      "learning_rate": 7.376141979066644e-06,
+      "loss": 0.5226,
+      "step": 1080
+    },
+    {
+      "epoch": 1.8956521739130436,
+      "grad_norm": 1.0538696180373945,
+      "learning_rate": 7.325378602983371e-06,
+      "loss": 0.4997,
+      "step": 1090
+    },
+    {
+      "epoch": 1.9130434782608696,
+      "grad_norm": 1.0941944221450914,
+      "learning_rate": 7.274307514336854e-06,
+      "loss": 0.5069,
+      "step": 1100
+    },
+    {
+      "epoch": 1.9304347826086956,
+      "grad_norm": 1.0475459053099843,
+      "learning_rate": 7.222935471259113e-06,
+      "loss": 0.4516,
+      "step": 1110
+    },
+    {
+      "epoch": 1.9478260869565216,
+      "grad_norm": 0.9438781712939313,
+      "learning_rate": 7.171269271706847e-06,
+      "loss": 0.4771,
+      "step": 1120
+    },
+    {
+      "epoch": 1.9652173913043478,
+      "grad_norm": 1.26652793242089,
+      "learning_rate": 7.119315752561879e-06,
+      "loss": 0.4628,
+      "step": 1130
+    },
+    {
+      "epoch": 1.982608695652174,
+      "grad_norm": 1.0230830545480807,
+      "learning_rate": 7.0670817887264375e-06,
+      "loss": 0.4961,
+      "step": 1140
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.3043673360211918,
+      "learning_rate": 7.014574292213423e-06,
+      "loss": 0.4854,
+      "step": 1150
+    },
+    {
+      "epoch": 2.017391304347826,
+      "grad_norm": 1.2839938386528644,
+      "learning_rate": 6.9618002112317525e-06,
+      "loss": 0.4159,
+      "step": 1160
+    },
+    {
+      "epoch": 2.034782608695652,
+      "grad_norm": 1.485302697617233,
+      "learning_rate": 6.908766529266915e-06,
+      "loss": 0.4513,
+      "step": 1170
+    },
+    {
+      "epoch": 2.0521739130434784,
+      "grad_norm": 1.3513626513686101,
+      "learning_rate": 6.855480264156864e-06,
+      "loss": 0.4147,
+      "step": 1180
+    },
+    {
+      "epoch": 2.0695652173913044,
+      "grad_norm": 1.1971012611734106,
+      "learning_rate": 6.8019484671633586e-06,
+      "loss": 0.4279,
+      "step": 1190
+    },
+    {
+      "epoch": 2.0869565217391304,
+      "grad_norm": 1.1493410270586013,
+      "learning_rate": 6.748178222038891e-06,
+      "loss": 0.4121,
+      "step": 1200
+    },
+    {
+      "epoch": 2.1043478260869564,
+      "grad_norm": 1.2061559921790461,
+      "learning_rate": 6.694176644089297e-06,
+      "loss": 0.4302,
+      "step": 1210
+    },
+    {
+      "epoch": 2.121739130434783,
+      "grad_norm": 1.2134461208204077,
+      "learning_rate": 6.639950879232218e-06,
+      "loss": 0.4143,
+      "step": 1220
+    },
+    {
+      "epoch": 2.139130434782609,
+      "grad_norm": 1.1936990104884397,
+      "learning_rate": 6.585508103051478e-06,
+      "loss": 0.411,
+      "step": 1230
+    },
+    {
+      "epoch": 2.1565217391304348,
+      "grad_norm": 1.473818496639146,
+      "learning_rate": 6.530855519847568e-06,
+      "loss": 0.4247,
+      "step": 1240
+    },
+    {
+      "epoch": 2.1739130434782608,
+      "grad_norm": 1.2273898491461919,
+      "learning_rate": 6.4760003616843106e-06,
+      "loss": 0.4,
+      "step": 1250
+    },
+    {
+      "epoch": 2.1913043478260867,
+      "grad_norm": 1.193058454958422,
+      "learning_rate": 6.420949887431855e-06,
+      "loss": 0.4016,
+      "step": 1260
+    },
+    {
+      "epoch": 2.208695652173913,
+      "grad_norm": 1.3961066733797438,
+      "learning_rate": 6.3657113818061225e-06,
+      "loss": 0.4408,
+      "step": 1270
+    },
+    {
+      "epoch": 2.226086956521739,
+      "grad_norm": 1.292360770899796,
+      "learning_rate": 6.310292154404844e-06,
+      "loss": 0.3999,
+      "step": 1280
+    },
+    {
+      "epoch": 2.243478260869565,
+      "grad_norm": 3.4982840024869417,
+      "learning_rate": 6.254699538740292e-06,
+      "loss": 0.4329,
+      "step": 1290
+    },
+    {
+      "epoch": 2.260869565217391,
+      "grad_norm": 2.265019151887774,
+      "learning_rate": 6.198940891268844e-06,
+      "loss": 0.4282,
+      "step": 1300
+    },
+    {
+      "epoch": 2.2782608695652176,
+      "grad_norm": 1.1398620538733453,
+      "learning_rate": 6.14302359041753e-06,
+      "loss": 0.4222,
+      "step": 1310
+    },
+    {
+      "epoch": 2.2956521739130435,
+      "grad_norm": 1.523968350586471,
+      "learning_rate": 6.086955035607655e-06,
+      "loss": 0.4317,
+      "step": 1320
+    },
+    {
+      "epoch": 2.3130434782608695,
+      "grad_norm": 2.2066066593878197,
+      "learning_rate": 6.030742646275647e-06,
+      "loss": 0.4212,
+      "step": 1330
+    },
+    {
+      "epoch": 2.3304347826086955,
+      "grad_norm": 1.2410016732543188,
+      "learning_rate": 5.9743938608912626e-06,
+      "loss": 0.4334,
+      "step": 1340
+    },
+    {
+      "epoch": 2.3478260869565215,
+      "grad_norm": 1.210595936193342,
+      "learning_rate": 5.917916135973263e-06,
+      "loss": 0.4153,
+      "step": 1350
+    },
+    {
+      "epoch": 2.365217391304348,
+      "grad_norm": 1.4742810962432356,
+      "learning_rate": 5.861316945102717e-06,
+      "loss": 0.4103,
+      "step": 1360
+    },
+    {
+      "epoch": 2.382608695652174,
+      "grad_norm": 1.1807404650454025,
+      "learning_rate": 5.804603777934032e-06,
+      "loss": 0.4104,
+      "step": 1370
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 1.32289460697371,
+      "learning_rate": 5.74778413920386e-06,
+      "loss": 0.4245,
+      "step": 1380
+    },
+    {
+      "epoch": 2.417391304347826,
+      "grad_norm": 1.4542008533192958,
+      "learning_rate": 5.690865547738021e-06,
+      "loss": 0.4412,
+      "step": 1390
+    },
+    {
+      "epoch": 2.4347826086956523,
+      "grad_norm": 1.2739934708683387,
+      "learning_rate": 5.6338555354565445e-06,
+      "loss": 0.4281,
+      "step": 1400
+    },
+    {
+      "epoch": 2.4521739130434783,
+      "grad_norm": 1.563662237308135,
+      "learning_rate": 5.576761646376987e-06,
+      "loss": 0.4359,
+      "step": 1410
+    },
+    {
+      "epoch": 2.4695652173913043,
+      "grad_norm": 1.225022509026475,
+      "learning_rate": 5.519591435616153e-06,
+      "loss": 0.4377,
+      "step": 1420
+    },
+    {
+      "epoch": 2.4869565217391303,
+      "grad_norm": 1.1686568195887326,
+      "learning_rate": 5.462352468390333e-06,
+      "loss": 0.4229,
+      "step": 1430
+    },
+    {
+      "epoch": 2.5043478260869563,
+      "grad_norm": 1.2575917217265098,
+      "learning_rate": 5.405052319014223e-06,
+      "loss": 0.4225,
+      "step": 1440
+    },
+    {
+      "epoch": 2.5217391304347827,
+      "grad_norm": 1.2880775613395745,
+      "learning_rate": 5.347698569898624e-06,
+      "loss": 0.4245,
+      "step": 1450
+    },
+    {
+      "epoch": 2.5391304347826087,
+      "grad_norm": 1.3156168488856514,
+      "learning_rate": 5.290298810547083e-06,
+      "loss": 0.3993,
+      "step": 1460
+    },
+    {
+      "epoch": 2.5565217391304347,
+      "grad_norm": 1.1736379125037195,
+      "learning_rate": 5.232860636551583e-06,
+      "loss": 0.4324,
+      "step": 1470
+    },
+    {
+      "epoch": 2.573913043478261,
+      "grad_norm": 1.2353830078230719,
+      "learning_rate": 5.175391648587443e-06,
+      "loss": 0.4185,
+      "step": 1480
+    },
+    {
+      "epoch": 2.591304347826087,
+      "grad_norm": 1.1458194395748447,
+      "learning_rate": 5.117899451407526e-06,
+      "loss": 0.4309,
+      "step": 1490
+    },
+    {
+      "epoch": 2.608695652173913,
+      "grad_norm": 1.2538290106691092,
+      "learning_rate": 5.060391652835925e-06,
+      "loss": 0.4292,
+      "step": 1500
+    },
+    {
+      "epoch": 2.626086956521739,
+      "grad_norm": 1.571430145629218,
+      "learning_rate": 5.002875862761234e-06,
+      "loss": 0.4268,
+      "step": 1510
+    },
+    {
+      "epoch": 2.643478260869565,
+      "grad_norm": 1.1840870452769912,
+      "learning_rate": 4.9453596921295435e-06,
+      "loss": 0.4319,
+      "step": 1520
+    },
+    {
+      "epoch": 2.660869565217391,
+      "grad_norm": 1.0371855886325287,
+      "learning_rate": 4.8878507519373055e-06,
+      "loss": 0.4378,
+      "step": 1530
+    },
+    {
+      "epoch": 2.6782608695652175,
+      "grad_norm": 1.0983294473272651,
+      "learning_rate": 4.830356652224181e-06,
+      "loss": 0.4058,
+      "step": 1540
+    },
+    {
+      "epoch": 2.6956521739130435,
+      "grad_norm": 1.127081289246426,
+      "learning_rate": 4.77288500106602e-06,
+      "loss": 0.4165,
+      "step": 1550
+    },
+    {
+      "epoch": 2.7130434782608694,
+      "grad_norm": 1.236794619724129,
+      "learning_rate": 4.715443403568103e-06,
+      "loss": 0.4148,
+      "step": 1560
+    },
+    {
+      "epoch": 2.730434782608696,
+      "grad_norm": 1.5345488197083625,
+      "learning_rate": 4.65803946085877e-06,
+      "loss": 0.4088,
+      "step": 1570
+    },
+    {
+      "epoch": 2.747826086956522,
+      "grad_norm": 1.3004765992472387,
+      "learning_rate": 4.600680769083585e-06,
+      "loss": 0.4359,
+      "step": 1580
+    },
+    {
+      "epoch": 2.765217391304348,
+      "grad_norm": 1.2501644106321015,
+      "learning_rate": 4.543374918400142e-06,
+      "loss": 0.4386,
+      "step": 1590
+    },
+    {
+      "epoch": 2.782608695652174,
+      "grad_norm": 1.3823645221856542,
+      "learning_rate": 4.486129491973687e-06,
+      "loss": 0.4517,
+      "step": 1600
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 1.2761252508293937,
+      "learning_rate": 4.4289520649736475e-06,
+      "loss": 0.4283,
+      "step": 1610
+    },
+    {
+      "epoch": 2.8173913043478263,
+      "grad_norm": 1.1632722443857555,
+      "learning_rate": 4.371850203571225e-06,
+      "loss": 0.421,
+      "step": 1620
+    },
+    {
+      "epoch": 2.8347826086956522,
+      "grad_norm": 1.2193532972111443,
+      "learning_rate": 4.314831463938184e-06,
+      "loss": 0.4188,
+      "step": 1630
+    },
+    {
+      "epoch": 2.8521739130434782,
+      "grad_norm": 1.4424575510136903,
+      "learning_rate": 4.257903391246954e-06,
+      "loss": 0.4094,
+      "step": 1640
+    },
+    {
+      "epoch": 2.869565217391304,
+      "grad_norm": 1.095166597880803,
+      "learning_rate": 4.201073518672195e-06,
+      "loss": 0.4411,
+      "step": 1650
+    },
+    {
+      "epoch": 2.8869565217391306,
+      "grad_norm": 1.2662835313537542,
+      "learning_rate": 4.144349366393949e-06,
+      "loss": 0.4423,
+      "step": 1660
+    },
+    {
+      "epoch": 2.9043478260869566,
+      "grad_norm": 1.1922956806962224,
+      "learning_rate": 4.08773844060251e-06,
+      "loss": 0.4254,
+      "step": 1670
+    },
+    {
+      "epoch": 2.9217391304347826,
+      "grad_norm": 1.257711038107532,
+      "learning_rate": 4.031248232505139e-06,
+      "loss": 0.4358,
+      "step": 1680
+    },
+    {
+      "epoch": 2.9391304347826086,
+      "grad_norm": 1.6971639546827986,
+      "learning_rate": 3.97488621733478e-06,
+      "loss": 0.4084,
+      "step": 1690
+    },
+    {
+      "epoch": 2.9565217391304346,
+      "grad_norm": 1.349596141789289,
+      "learning_rate": 3.918659853360864e-06,
+      "loss": 0.4274,
+      "step": 1700
+    },
+    {
+      "epoch": 2.973913043478261,
+      "grad_norm": 1.3678548460193103,
+      "learning_rate": 3.862576580902383e-06,
+      "loss": 0.4187,
+      "step": 1710
+    },
+    {
+      "epoch": 2.991304347826087,
+      "grad_norm": 1.2860411793928146,
+      "learning_rate": 3.8066438213433234e-06,
+      "loss": 0.4303,
+      "step": 1720
+    },
+    {
+      "epoch": 3.008695652173913,
+      "grad_norm": 1.5134066645346198,
+      "learning_rate": 3.7508689761506055e-06,
+      "loss": 0.3752,
+      "step": 1730
+    },
+    {
+      "epoch": 3.026086956521739,
+      "grad_norm": 1.424990341239896,
+      "learning_rate": 3.6952594258946693e-06,
+      "loss": 0.3434,
+      "step": 1740
+    },
+    {
+      "epoch": 3.0434782608695654,
+      "grad_norm": 1.6649951275575265,
+      "learning_rate": 3.6398225292728185e-06,
+      "loss": 0.3607,
+      "step": 1750
+    },
+    {
+      "epoch": 3.0608695652173914,
+      "grad_norm": 1.8127043018682598,
+      "learning_rate": 3.584565622135453e-06,
+      "loss": 0.3623,
+      "step": 1760
+    },
+    {
+      "epoch": 3.0782608695652174,
+      "grad_norm": 1.3226724796309823,
+      "learning_rate": 3.5294960165153363e-06,
+      "loss": 0.3579,
+      "step": 1770
+    },
+    {
+      "epoch": 3.0956521739130434,
+      "grad_norm": 1.4747389958766943,
+      "learning_rate": 3.474620999660007e-06,
+      "loss": 0.3973,
+      "step": 1780
+    },
+    {
+      "epoch": 3.1130434782608694,
+      "grad_norm": 1.9045203788068308,
+      "learning_rate": 3.4199478330674745e-06,
+      "loss": 0.3579,
+      "step": 1790
+    },
+    {
+      "epoch": 3.130434782608696,
+      "grad_norm": 1.2899295921725031,
+      "learning_rate": 3.365483751525317e-06,
+      "loss": 0.3634,
+      "step": 1800
+    },
+    {
+      "epoch": 3.1478260869565218,
+      "grad_norm": 1.4141207800877191,
+      "learning_rate": 3.3112359621533193e-06,
+      "loss": 0.3427,
+      "step": 1810
+    },
+    {
+      "epoch": 3.1652173913043478,
+      "grad_norm": 1.4287974659319675,
+      "learning_rate": 3.257211643449768e-06,
+      "loss": 0.3425,
+      "step": 1820
+    },
+    {
+      "epoch": 3.1826086956521737,
+      "grad_norm": 1.3872599954983553,
+      "learning_rate": 3.203417944341536e-06,
+      "loss": 0.3458,
+      "step": 1830
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 2.2293598598474245,
+      "learning_rate": 3.149861983238082e-06,
+      "loss": 0.3689,
+      "step": 1840
+    },
+    {
+      "epoch": 3.217391304347826,
+      "grad_norm": 2.0325771541464217,
+      "learning_rate": 3.0965508470894812e-06,
+      "loss": 0.3521,
+      "step": 1850
+    },
+    {
+      "epoch": 3.234782608695652,
+      "grad_norm": 1.4460102517208502,
+      "learning_rate": 3.0434915904486284e-06,
+      "loss": 0.358,
+      "step": 1860
+    },
+    {
+      "epoch": 3.252173913043478,
+      "grad_norm": 1.3048211800924494,
+      "learning_rate": 2.990691234537721e-06,
+      "loss": 0.3405,
+      "step": 1870
+    },
+    {
+      "epoch": 3.269565217391304,
+      "grad_norm": 1.3241574204938475,
+      "learning_rate": 2.938156766319156e-06,
+      "loss": 0.384,
+      "step": 1880
+    },
+    {
+      "epoch": 3.2869565217391306,
+      "grad_norm": 1.6150241974779345,
+      "learning_rate": 2.885895137570958e-06,
+      "loss": 0.3621,
+      "step": 1890
+    },
+    {
+      "epoch": 3.3043478260869565,
+      "grad_norm": 1.2579063705669289,
+      "learning_rate": 2.83391326396687e-06,
+      "loss": 0.3884,
+      "step": 1900
+    },
+    {
+      "epoch": 3.3217391304347825,
+      "grad_norm": 1.3801202825398762,
+      "learning_rate": 2.7822180241612077e-06,
+      "loss": 0.3798,
+      "step": 1910
+    },
+    {
+      "epoch": 3.3391304347826085,
+      "grad_norm": 1.396049263670027,
+      "learning_rate": 2.7308162588786303e-06,
+      "loss": 0.3605,
+      "step": 1920
+    },
+    {
+      "epoch": 3.356521739130435,
+      "grad_norm": 1.3748023066384665,
+      "learning_rate": 2.6797147700089167e-06,
+      "loss": 0.3579,
+      "step": 1930
+    },
+    {
+      "epoch": 3.373913043478261,
+      "grad_norm": 1.5175283161299356,
+      "learning_rate": 2.6289203197068834e-06,
+      "loss": 0.3784,
+      "step": 1940
+    },
+    {
+      "epoch": 3.391304347826087,
+      "grad_norm": 1.5851194282172685,
+      "learning_rate": 2.5784396294975677e-06,
+      "loss": 0.3606,
+      "step": 1950
+    },
+    {
+      "epoch": 3.408695652173913,
+      "grad_norm": 1.349779292865959,
+      "learning_rate": 2.528279379386783e-06,
+      "loss": 0.3694,
+      "step": 1960
+    },
+    {
+      "epoch": 3.426086956521739,
+      "grad_norm": 1.5339986272397241,
+      "learning_rate": 2.478446206977159e-06,
+      "loss": 0.3771,
+      "step": 1970
+    },
+    {
+      "epoch": 3.4434782608695653,
+      "grad_norm": 1.430106842741753,
+      "learning_rate": 2.4289467065898085e-06,
+      "loss": 0.369,
+      "step": 1980
+    },
+    {
+      "epoch": 3.4608695652173913,
+      "grad_norm": 1.457323252888418,
+      "learning_rate": 2.3797874283917127e-06,
+      "loss": 0.3403,
+      "step": 1990
+    },
+    {
+      "epoch": 3.4782608695652173,
+      "grad_norm": 1.4282143674816088,
+      "learning_rate": 2.3309748775289497e-06,
+      "loss": 0.3845,
+      "step": 2000
+    },
+    {
+      "epoch": 3.4956521739130437,
+      "grad_norm": 1.3173625440079775,
+      "learning_rate": 2.282515513265885e-06,
+      "loss": 0.3695,
+      "step": 2010
+    },
+    {
+      "epoch": 3.5130434782608697,
+      "grad_norm": 1.5391795918668048,
+      "learning_rate": 2.2344157481304267e-06,
+      "loss": 0.384,
+      "step": 2020
+    },
+    {
+      "epoch": 3.5304347826086957,
+      "grad_norm": 1.5462876476217569,
+      "learning_rate": 2.1866819470654727e-06,
+      "loss": 0.362,
+      "step": 2030
+    },
+    {
+      "epoch": 3.5478260869565217,
+      "grad_norm": 1.5819871351937547,
+      "learning_rate": 2.1393204265866467e-06,
+      "loss": 0.3788,
+      "step": 2040
+    },
+    {
+      "epoch": 3.5652173913043477,
+      "grad_norm": 1.6521343485618902,
+      "learning_rate": 2.09233745394645e-06,
+      "loss": 0.3785,
+      "step": 2050
+    },
+    {
+      "epoch": 3.5826086956521737,
+      "grad_norm": 1.5040618975765006,
+      "learning_rate": 2.0457392463049285e-06,
+      "loss": 0.359,
+      "step": 2060
+    },
+    {
+      "epoch": 3.6,
+      "grad_norm": 1.6074930743024214,
+      "learning_rate": 1.9995319699069664e-06,
+      "loss": 0.3652,
+      "step": 2070
+    },
+    {
+      "epoch": 3.617391304347826,
+      "grad_norm": 1.5631144357601194,
+      "learning_rate": 1.9537217392663218e-06,
+      "loss": 0.3626,
+      "step": 2080
+    },
+    {
+      "epoch": 3.634782608695652,
+      "grad_norm": 1.5044302087614625,
+      "learning_rate": 1.908314616356505e-06,
+      "loss": 0.388,
+      "step": 2090
+    },
+    {
+      "epoch": 3.6521739130434785,
+      "grad_norm": 1.3802858769348085,
+      "learning_rate": 1.8633166098086103e-06,
+      "loss": 0.3629,
+      "step": 2100
+    },
+    {
+      "epoch": 3.6695652173913045,
+      "grad_norm": 1.4222636474825245,
+      "learning_rate": 1.818733674116207e-06,
+      "loss": 0.3651,
+      "step": 2110
+    },
+    {
+      "epoch": 3.6869565217391305,
+      "grad_norm": 1.5026251469181622,
+      "learning_rate": 1.7745717088473895e-06,
+      "loss": 0.3772,
+      "step": 2120
+    },
+    {
+      "epoch": 3.7043478260869565,
+      "grad_norm": 1.4848721046911413,
+      "learning_rate": 1.7308365578641089e-06,
+      "loss": 0.3522,
+      "step": 2130
+    },
+    {
+      "epoch": 3.7217391304347824,
+      "grad_norm": 1.6975508268605253,
+      "learning_rate": 1.687534008548854e-06,
+      "loss": 0.3775,
+      "step": 2140
+    },
+    {
+      "epoch": 3.7391304347826084,
+      "grad_norm": 1.7810729043290516,
+      "learning_rate": 1.6446697910388294e-06,
+      "loss": 0.3647,
+      "step": 2150
+    },
+    {
+      "epoch": 3.756521739130435,
+      "grad_norm": 1.4737138340480456,
+      "learning_rate": 1.6022495774676916e-06,
+      "loss": 0.3723,
+      "step": 2160
+    },
+    {
+      "epoch": 3.773913043478261,
+      "grad_norm": 1.5559891939482013,
+      "learning_rate": 1.5602789812149727e-06,
+      "loss": 0.3503,
+      "step": 2170
+    },
+    {
+      "epoch": 3.791304347826087,
+      "grad_norm": 1.663018576508128,
+      "learning_rate": 1.5187635561632685e-06,
+      "loss": 0.3561,
+      "step": 2180
+    },
+    {
+      "epoch": 3.8086956521739133,
+      "grad_norm": 2.634784063202797,
+      "learning_rate": 1.477708795963308e-06,
+      "loss": 0.3406,
+      "step": 2190
+    },
+    {
+      "epoch": 3.8260869565217392,
+      "grad_norm": 1.5139630905983323,
+      "learning_rate": 1.4371201333069868e-06,
+      "loss": 0.3681,
+      "step": 2200
+    },
+    {
+      "epoch": 3.8434782608695652,
+      "grad_norm": 2.3176470378092424,
+      "learning_rate": 1.3970029392084771e-06,
+      "loss": 0.3541,
+      "step": 2210
+    },
+    {
+      "epoch": 3.860869565217391,
+      "grad_norm": 1.5959870558495743,
+      "learning_rate": 1.3573625222934829e-06,
+      "loss": 0.3675,
+      "step": 2220
+    },
+    {
+      "epoch": 3.878260869565217,
+      "grad_norm": 1.6337373914467612,
+      "learning_rate": 1.3182041280967656e-06,
+      "loss": 0.3514,
+      "step": 2230
+    },
+    {
+      "epoch": 3.8956521739130436,
+      "grad_norm": 1.4395314046176468,
+      "learning_rate": 1.2795329383680138e-06,
+      "loss": 0.3566,
+      "step": 2240
+    },
+    {
+      "epoch": 3.9130434782608696,
+      "grad_norm": 1.3199487970653823,
+      "learning_rate": 1.241354070386151e-06,
+      "loss": 0.3441,
+      "step": 2250
+    },
+    {
+      "epoch": 3.9304347826086956,
+      "grad_norm": 1.7513753951221092,
+      "learning_rate": 1.2036725762821783e-06,
+      "loss": 0.3628,
+      "step": 2260
+    },
+    {
+      "epoch": 3.9478260869565216,
+      "grad_norm": 1.5442153725537786,
+      "learning_rate": 1.1664934423706348e-06,
+      "loss": 0.3497,
+      "step": 2270
+    },
+    {
+      "epoch": 3.965217391304348,
+      "grad_norm": 1.663632062954429,
+      "learning_rate": 1.12982158848977e-06,
+      "loss": 0.3576,
+      "step": 2280
+    },
+    {
+      "epoch": 3.982608695652174,
+      "grad_norm": 1.8278674162694661,
+      "learning_rate": 1.0936618673505112e-06,
+      "loss": 0.3744,
+      "step": 2290
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.4687537010366873,
+      "learning_rate": 1.0580190638943138e-06,
+      "loss": 0.3549,
+      "step": 2300
+    },
+    {
+      "epoch": 4.017391304347826,
+      "grad_norm": 1.6716521197547283,
+      "learning_rate": 1.022897894659981e-06,
+      "loss": 0.3487,
+      "step": 2310
+    },
+    {
+      "epoch": 4.034782608695652,
+      "grad_norm": 1.4645341272970525,
+      "learning_rate": 9.883030071595335e-07,
+      "loss": 0.3158,
+      "step": 2320
+    },
+    {
+      "epoch": 4.052173913043478,
+      "grad_norm": 1.6557398280558575,
+      "learning_rate": 9.542389792632112e-07,
+      "loss": 0.331,
+      "step": 2330
+    },
+    {
+      "epoch": 4.069565217391304,
+      "grad_norm": 1.5960738974574165,
+      "learning_rate": 9.20710318593701e-07,
+      "loss": 0.3265,
+      "step": 2340
+    },
+    {
+      "epoch": 4.086956521739131,
+      "grad_norm": 1.4988502007449698,
+      "learning_rate": 8.877214619296421e-07,
+      "loss": 0.336,
+      "step": 2350
+    },
+    {
+      "epoch": 4.104347826086957,
+      "grad_norm": 1.4424789164433198,
+      "learning_rate": 8.552767746185215e-07,
+      "loss": 0.3347,
+      "step": 2360
+    },
+    {
+      "epoch": 4.121739130434783,
+      "grad_norm": 2.2719026550199475,
+      "learning_rate": 8.233805499990166e-07,
+      "loss": 0.3305,
+      "step": 2370
+    },
+    {
+      "epoch": 4.139130434782609,
+      "grad_norm": 1.4765476496178676,
+      "learning_rate": 7.920370088328672e-07,
+      "loss": 0.3484,
+      "step": 2380
+    },
+    {
+      "epoch": 4.156521739130435,
+      "grad_norm": 1.7030261009934748,
+      "learning_rate": 7.612502987463477e-07,
+      "loss": 0.3163,
+      "step": 2390
+    },
+    {
+      "epoch": 4.173913043478261,
+      "grad_norm": 1.4010667448317558,
+      "learning_rate": 7.310244936814232e-07,
+      "loss": 0.3222,
+      "step": 2400
+    },
+    {
+      "epoch": 4.191304347826087,
+      "grad_norm": 1.381057060755463,
+      "learning_rate": 7.013635933566515e-07,
+      "loss": 0.3175,
+      "step": 2410
+    },
+    {
+      "epoch": 4.208695652173913,
+      "grad_norm": 1.6489840526877189,
+      "learning_rate": 6.72271522737909e-07,
+      "loss": 0.3027,
+      "step": 2420
+    },
+    {
+      "epoch": 4.226086956521739,
+      "grad_norm": 1.4490054494783193,
+      "learning_rate": 6.437521315190087e-07,
+      "loss": 0.3529,
+      "step": 2430
+    },
+    {
+      "epoch": 4.243478260869566,
+      "grad_norm": 1.2850499028342035,
+      "learning_rate": 6.158091936122773e-07,
+      "loss": 0.3285,
+      "step": 2440
+    },
+    {
+      "epoch": 4.260869565217392,
+      "grad_norm": 1.2958305173488132,
+      "learning_rate": 5.884464066491613e-07,
+      "loss": 0.3128,
+      "step": 2450
+    },
+    {
+      "epoch": 4.278260869565218,
+      "grad_norm": 1.716194835927619,
+      "learning_rate": 5.616673914909282e-07,
+      "loss": 0.3377,
+      "step": 2460
+    },
+    {
+      "epoch": 4.2956521739130435,
+      "grad_norm": 1.5456459782523917,
+      "learning_rate": 5.354756917495224e-07,
+      "loss": 0.359,
+      "step": 2470
+    },
+    {
+      "epoch": 4.3130434782608695,
+      "grad_norm": 1.7049849318119408,
+      "learning_rate": 5.098747733186498e-07,
+      "loss": 0.3194,
+      "step": 2480
+    },
+    {
+      "epoch": 4.3304347826086955,
+      "grad_norm": 1.4376855175332917,
+      "learning_rate": 4.84868023915141e-07,
+      "loss": 0.3296,
+      "step": 2490
+    },
+    {
+      "epoch": 4.3478260869565215,
+      "grad_norm": 1.5575044356910057,
+      "learning_rate": 4.6045875263066474e-07,
+      "loss": 0.3347,
+      "step": 2500
+    },
+    {
+      "epoch": 4.3652173913043475,
+      "grad_norm": 1.6696231773916848,
+      "learning_rate": 4.366501894938363e-07,
+      "loss": 0.3509,
+      "step": 2510
+    },
+    {
+      "epoch": 4.3826086956521735,
+      "grad_norm": 1.4793874890497716,
+      "learning_rate": 4.1344548504280213e-07,
+      "loss": 0.316,
+      "step": 2520
+    },
+    {
+      "epoch": 4.4,
+      "grad_norm": 1.7504184745527158,
+      "learning_rate": 3.908477099083291e-07,
+      "loss": 0.3368,
+      "step": 2530
+    },
+    {
+      "epoch": 4.417391304347826,
+      "grad_norm": 1.5433746257360492,
+      "learning_rate": 3.6885985440747895e-07,
+      "loss": 0.3147,
+      "step": 2540
+    },
+    {
+      "epoch": 4.434782608695652,
+      "grad_norm": 1.6205451434238438,
+      "learning_rate": 3.474848281479032e-07,
+      "loss": 0.3198,
+      "step": 2550
+    },
+    {
+      "epoch": 4.452173913043478,
+      "grad_norm": 1.7748707264655563,
+      "learning_rate": 3.2672545964282263e-07,
+      "loss": 0.3337,
+      "step": 2560
+    },
+    {
+      "epoch": 4.469565217391304,
+      "grad_norm": 2.034670741566414,
+      "learning_rate": 3.0658449593673256e-07,
+      "loss": 0.2871,
+      "step": 2570
+    },
+    {
+      "epoch": 4.48695652173913,
+      "grad_norm": 1.9535751002955573,
+      "learning_rate": 2.8706460224189656e-07,
+      "loss": 0.3106,
+      "step": 2580
+    },
+    {
+      "epoch": 4.504347826086956,
+      "grad_norm": 1.4860584696429842,
+      "learning_rate": 2.6816836158566117e-07,
+      "loss": 0.3304,
+      "step": 2590
+    },
+    {
+      "epoch": 4.521739130434782,
+      "grad_norm": 1.5646385009356472,
+      "learning_rate": 2.498982744686501e-07,
+      "loss": 0.328,
+      "step": 2600
+    },
+    {
+      "epoch": 4.539130434782608,
+      "grad_norm": 1.6223923163321592,
+      "learning_rate": 2.3225675853387974e-07,
+      "loss": 0.3099,
+      "step": 2610
+    },
+    {
+      "epoch": 4.556521739130435,
+      "grad_norm": 1.4069397141672941,
+      "learning_rate": 2.1524614824683377e-07,
+      "loss": 0.3318,
+      "step": 2620
+    },
+    {
+      "epoch": 4.573913043478261,
+      "grad_norm": 1.5612850701477952,
+      "learning_rate": 1.98868694586552e-07,
+      "loss": 0.3376,
+      "step": 2630
+    },
+    {
+      "epoch": 4.591304347826087,
+      "grad_norm": 1.6586621274603865,
+      "learning_rate": 1.8312656474776093e-07,
+      "loss": 0.3034,
+      "step": 2640
+    },
+    {
+      "epoch": 4.608695652173913,
+      "grad_norm": 1.6277894314324821,
+      "learning_rate": 1.6802184185409355e-07,
+      "loss": 0.334,
+      "step": 2650
+    },
+    {
+      "epoch": 4.626086956521739,
+      "grad_norm": 1.5367253210879437,
+      "learning_rate": 1.5355652468243332e-07,
+      "loss": 0.3368,
+      "step": 2660
+    },
+    {
+      "epoch": 4.643478260869565,
+      "grad_norm": 1.42580964043298,
+      "learning_rate": 1.3973252739842236e-07,
+      "loss": 0.3107,
+      "step": 2670
+    },
+    {
+      "epoch": 4.660869565217391,
+      "grad_norm": 1.5836881272840495,
+      "learning_rate": 1.2655167930316236e-07,
+      "loss": 0.3337,
+      "step": 2680
+    },
+    {
+      "epoch": 4.678260869565217,
+      "grad_norm": 1.3933170878068741,
+      "learning_rate": 1.1401572459114441e-07,
+      "loss": 0.2988,
+      "step": 2690
+    },
+    {
+      "epoch": 4.695652173913043,
+      "grad_norm": 1.6629891174849243,
+      "learning_rate": 1.0212632211944906e-07,
+      "loss": 0.3038,
+      "step": 2700
+    },
+    {
+      "epoch": 4.71304347826087,
+      "grad_norm": 1.5180536343786482,
+      "learning_rate": 9.088504518822817e-08,
+      "loss": 0.2919,
+      "step": 2710
+    },
+    {
+      "epoch": 4.730434782608696,
+      "grad_norm": 1.4774316548557718,
+      "learning_rate": 8.029338133251518e-08,
+      "loss": 0.3253,
+      "step": 2720
+    },
+    {
+      "epoch": 4.747826086956522,
+      "grad_norm": 1.4916465022440153,
+      "learning_rate": 7.035273212538274e-08,
+      "loss": 0.3346,
+      "step": 2730
+    },
+    {
+      "epoch": 4.765217391304348,
+      "grad_norm": 1.6408617577143456,
+      "learning_rate": 6.10644129924759e-08,
+      "loss": 0.2921,
+      "step": 2740
+    },
+    {
+      "epoch": 4.782608695652174,
+      "grad_norm": 1.5935791526940084,
+      "learning_rate": 5.242965303794312e-08,
+      "loss": 0.3298,
+      "step": 2750
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 1.3615616479228225,
+      "learning_rate": 4.444959488179301e-08,
+      "loss": 0.3084,
+      "step": 2760
+    },
+    {
+      "epoch": 4.817391304347826,
+      "grad_norm": 1.3883499339712515,
+      "learning_rate": 3.7125294508693066e-08,
+      "loss": 0.3117,
+      "step": 2770
+    },
+    {
+      "epoch": 4.834782608695652,
+      "grad_norm": 1.483718903196712,
+      "learning_rate": 3.045772112823253e-08,
+      "loss": 0.3259,
+      "step": 2780
+    },
+    {
+      "epoch": 4.852173913043478,
+      "grad_norm": 1.5939760050066456,
+      "learning_rate": 2.4447757046670017e-08,
+      "loss": 0.3079,
+      "step": 2790
+    },
+    {
+      "epoch": 4.869565217391305,
+      "grad_norm": 1.5266765004604943,
+      "learning_rate": 1.9096197550179664e-08,
+      "loss": 0.3214,
+      "step": 2800
+    },
+    {
+      "epoch": 4.886956521739131,
+      "grad_norm": 1.7078593350970195,
+      "learning_rate": 1.4403750799613092e-08,
+      "loss": 0.3419,
+      "step": 2810
+    },
+    {
+      "epoch": 4.904347826086957,
+      "grad_norm": 1.595487144492393,
+      "learning_rate": 1.0371037736787714e-08,
+      "loss": 0.3224,
+      "step": 2820
+    },
+    {
+      "epoch": 4.921739130434783,
+      "grad_norm": 1.470874925227166,
+      "learning_rate": 6.998592002321336e-09,
+      "loss": 0.3303,
+      "step": 2830
+    },
+    {
+      "epoch": 4.939130434782609,
+      "grad_norm": 1.6419169053737916,
+      "learning_rate": 4.286859865014204e-09,
+      "loss": 0.3237,
+      "step": 2840
+    },
+    {
+      "epoch": 4.956521739130435,
+      "grad_norm": 1.7910736800765954,
+      "learning_rate": 2.236200162798463e-09,
+      "loss": 0.2943,
+      "step": 2850
+    },
+    {
+      "epoch": 4.973913043478261,
+      "grad_norm": 1.4483382289129645,
+      "learning_rate": 8.468842552505907e-10,
+      "loss": 0.3202,
+      "step": 2860
+    },
+    {
+      "epoch": 4.9913043478260875,
+      "grad_norm": 1.4884875727469478,
+      "learning_rate": 1.1909598768400366e-10,
+      "loss": 0.3389,
+      "step": 2870
+    },
+    {
+      "epoch": 5.0,
+      "step": 2875,
+      "total_flos": 33447808868352.0,
+      "train_loss": 0.42618958398570184,
+      "train_runtime": 1921.5134,
+      "train_samples_per_second": 11.957,
+      "train_steps_per_second": 1.496
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2875,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 10000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 33447808868352.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ce5cae7710446ef8c6957d71b42b5f7d1d0a240b0d25cb8116775791199f000
+size 7953