Upload folder using huggingface_hub

.gitattributes

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

added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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
+}

config.json

@@ -0,0 +1,38 @@
+{
+  "architectures": [
+    "Qwen3MoeForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "decoder_sparse_step": 1,
+  "eos_token_id": 151643,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 6144,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 48,
+  "mlp_only_layers": [],
+  "model_type": "qwen3_moe",
+  "moe_intermediate_size": 768,
+  "norm_topk_prob": true,
+  "num_attention_heads": 32,
+  "num_experts": 128,
+  "num_experts_per_tok": 8,
+  "num_hidden_layers": 48,
+  "num_key_value_heads": 4,
+  "output_router_logits": false,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "router_aux_loss_coef": 0.001,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.51.3"
+}

model-00001-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdcd0659d5c83f69fd221a54127ce14bed7726dbc0ef0a2976f32c3073d6026c
+size 4997184968

model-00002-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72aceff497063fe83cb3dc3d3e41e86a667f4d592fe441d30e6a16bf9e40a07c
+size 4997741608

model-00003-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e3bd8583b71aae2e18111230c12dbdd40cf7732926a2d11266da50df04390d3
+size 4997742208

model-00004-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21f549273bf9369c976a5bab2ecc5959d5945b64356ed888263c38220072a02d
+size 4997743184

model-00005-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4280639b7b4ae9c8874de6ae43178a53389d2680892f411fa746d034435bdaa2
+size 4997743184

model-00006-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d55679826d9f2e2b1dae028e2304c3852c692cc7a1dbe7f2a90792ca7739e833
+size 4997743184

model-00007-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3857c935f5890e8425303c7a0d7279b371304db607b0fe1986d7f045c5d20241
+size 4997743184

model-00008-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:729e1e4f5e0ef974add32723d94f6c01f94eefa862c26360be3aac3bac5274c6
+size 4997743184

model-00009-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9900f95f98c1d3ddc35bc63337d3ca4ab600f137facef85357eef0baee89e446
+size 4997743184

model-00010-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53420444113f2e98b433ad1530ce6e45a9265f2d00fb2b0d1ca6fca2c463af57
+size 4997743184

model-00011-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:974c8c38dc7111cfe676b7af88b0d0fde89afd27c537c602a8810df1e3d6e449
+size 4997743184

model-00012-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:196d51137c91dccf881d210581859e4e463264b5ea07596abd8b13857648c7e4
+size 4997743184

model-00013-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd8223a61ae9a11c994c91900914e57ef816a8ee15dc755a95d68afcd1fe9578
+size 1094220288

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:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

tokenizer_config.json

@@ -0,0 +1,241 @@
+{
+  "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
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set content = message.content %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in message.content %}\n                {%- set content = message.content.split('</think>')[-1].lstrip('\\n') %}\n                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
+  "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
+}

trainer_state.json

@@ -0,0 +1,2134 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.5420054200542005,
+  "eval_steps": 500,
+  "global_step": 300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0018066847335140017,
+      "grad_norm": 2.8401210755443786,
+      "learning_rate": 0.0,
+      "loss": 0.8251,
+      "step": 1
+    },
+    {
+      "epoch": 0.0036133694670280035,
+      "grad_norm": 2.884596771811998,
+      "learning_rate": 1.7857142857142857e-06,
+      "loss": 0.8284,
+      "step": 2
+    },
+    {
+      "epoch": 0.005420054200542005,
+      "grad_norm": 2.885426730353446,
+      "learning_rate": 3.5714285714285714e-06,
+      "loss": 0.8426,
+      "step": 3
+    },
+    {
+      "epoch": 0.007226738934056007,
+      "grad_norm": 2.651910724864975,
+      "learning_rate": 5.357142857142857e-06,
+      "loss": 0.8329,
+      "step": 4
+    },
+    {
+      "epoch": 0.009033423667570008,
+      "grad_norm": 2.138099539200911,
+      "learning_rate": 7.142857142857143e-06,
+      "loss": 0.8087,
+      "step": 5
+    },
+    {
+      "epoch": 0.01084010840108401,
+      "grad_norm": 1.5516804515831568,
+      "learning_rate": 8.92857142857143e-06,
+      "loss": 0.7878,
+      "step": 6
+    },
+    {
+      "epoch": 0.012646793134598013,
+      "grad_norm": 1.3537216088647457,
+      "learning_rate": 1.0714285714285714e-05,
+      "loss": 0.76,
+      "step": 7
+    },
+    {
+      "epoch": 0.014453477868112014,
+      "grad_norm": 2.2798033166818095,
+      "learning_rate": 1.25e-05,
+      "loss": 0.7502,
+      "step": 8
+    },
+    {
+      "epoch": 0.016260162601626018,
+      "grad_norm": 2.4280919600136315,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 0.7449,
+      "step": 9
+    },
+    {
+      "epoch": 0.018066847335140017,
+      "grad_norm": 1.931081683969551,
+      "learning_rate": 1.6071428571428572e-05,
+      "loss": 0.7271,
+      "step": 10
+    },
+    {
+      "epoch": 0.01987353206865402,
+      "grad_norm": 2.033957326273334,
+      "learning_rate": 1.785714285714286e-05,
+      "loss": 0.7224,
+      "step": 11
+    },
+    {
+      "epoch": 0.02168021680216802,
+      "grad_norm": 1.7440303482576578,
+      "learning_rate": 1.9642857142857145e-05,
+      "loss": 0.7373,
+      "step": 12
+    },
+    {
+      "epoch": 0.023486901535682024,
+      "grad_norm": 1.1423427090231095,
+      "learning_rate": 2.1428571428571428e-05,
+      "loss": 0.7033,
+      "step": 13
+    },
+    {
+      "epoch": 0.025293586269196026,
+      "grad_norm": 0.9350758811837806,
+      "learning_rate": 2.3214285714285715e-05,
+      "loss": 0.6886,
+      "step": 14
+    },
+    {
+      "epoch": 0.02710027100271003,
+      "grad_norm": 0.9041529205527444,
+      "learning_rate": 2.5e-05,
+      "loss": 0.6858,
+      "step": 15
+    },
+    {
+      "epoch": 0.028906955736224028,
+      "grad_norm": 0.7774159470583337,
+      "learning_rate": 2.6785714285714288e-05,
+      "loss": 0.676,
+      "step": 16
+    },
+    {
+      "epoch": 0.03071364046973803,
+      "grad_norm": 0.6642849851627606,
+      "learning_rate": 2.857142857142857e-05,
+      "loss": 0.6816,
+      "step": 17
+    },
+    {
+      "epoch": 0.032520325203252036,
+      "grad_norm": 0.6696862181877565,
+      "learning_rate": 3.0357142857142857e-05,
+      "loss": 0.6727,
+      "step": 18
+    },
+    {
+      "epoch": 0.03432700993676603,
+      "grad_norm": 0.6199596133483118,
+      "learning_rate": 3.2142857142857144e-05,
+      "loss": 0.6678,
+      "step": 19
+    },
+    {
+      "epoch": 0.036133694670280034,
+      "grad_norm": 0.5724104094717255,
+      "learning_rate": 3.392857142857143e-05,
+      "loss": 0.6448,
+      "step": 20
+    },
+    {
+      "epoch": 0.037940379403794036,
+      "grad_norm": 0.5762214978597714,
+      "learning_rate": 3.571428571428572e-05,
+      "loss": 0.6506,
+      "step": 21
+    },
+    {
+      "epoch": 0.03974706413730804,
+      "grad_norm": 0.5444644294356963,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.6413,
+      "step": 22
+    },
+    {
+      "epoch": 0.04155374887082204,
+      "grad_norm": 0.4801008888914334,
+      "learning_rate": 3.928571428571429e-05,
+      "loss": 0.6425,
+      "step": 23
+    },
+    {
+      "epoch": 0.04336043360433604,
+      "grad_norm": 0.4780671650041637,
+      "learning_rate": 4.107142857142857e-05,
+      "loss": 0.6489,
+      "step": 24
+    },
+    {
+      "epoch": 0.045167118337850046,
+      "grad_norm": 0.5145358853730851,
+      "learning_rate": 4.2857142857142856e-05,
+      "loss": 0.6306,
+      "step": 25
+    },
+    {
+      "epoch": 0.04697380307136405,
+      "grad_norm": 0.4229563893889767,
+      "learning_rate": 4.464285714285715e-05,
+      "loss": 0.6453,
+      "step": 26
+    },
+    {
+      "epoch": 0.04878048780487805,
+      "grad_norm": 0.4021332182222708,
+      "learning_rate": 4.642857142857143e-05,
+      "loss": 0.6251,
+      "step": 27
+    },
+    {
+      "epoch": 0.05058717253839205,
+      "grad_norm": 0.48615223659558016,
+      "learning_rate": 4.8214285714285716e-05,
+      "loss": 0.6272,
+      "step": 28
+    },
+    {
+      "epoch": 0.052393857271906055,
+      "grad_norm": 0.46304419453924084,
+      "learning_rate": 5e-05,
+      "loss": 0.6385,
+      "step": 29
+    },
+    {
+      "epoch": 0.05420054200542006,
+      "grad_norm": 0.4055914485047887,
+      "learning_rate": 4.999955240022902e-05,
+      "loss": 0.6406,
+      "step": 30
+    },
+    {
+      "epoch": 0.05600722673893405,
+      "grad_norm": 0.4320919468210144,
+      "learning_rate": 4.999820961694372e-05,
+      "loss": 0.6303,
+      "step": 31
+    },
+    {
+      "epoch": 0.057813911472448055,
+      "grad_norm": 0.4042983593828653,
+      "learning_rate": 4.999597169822646e-05,
+      "loss": 0.6186,
+      "step": 32
+    },
+    {
+      "epoch": 0.05962059620596206,
+      "grad_norm": 0.4310835292518631,
+      "learning_rate": 4.9992838724212585e-05,
+      "loss": 0.6287,
+      "step": 33
+    },
+    {
+      "epoch": 0.06142728093947606,
+      "grad_norm": 0.412271814827728,
+      "learning_rate": 4.9988810807087584e-05,
+      "loss": 0.6165,
+      "step": 34
+    },
+    {
+      "epoch": 0.06323396567299007,
+      "grad_norm": 0.40598837987994935,
+      "learning_rate": 4.998388809108303e-05,
+      "loss": 0.622,
+      "step": 35
+    },
+    {
+      "epoch": 0.06504065040650407,
+      "grad_norm": 0.4223908220324365,
+      "learning_rate": 4.997807075247146e-05,
+      "loss": 0.6189,
+      "step": 36
+    },
+    {
+      "epoch": 0.06684733514001806,
+      "grad_norm": 0.38197899028265064,
+      "learning_rate": 4.997135899956001e-05,
+      "loss": 0.623,
+      "step": 37
+    },
+    {
+      "epoch": 0.06865401987353206,
+      "grad_norm": 0.3885330150920031,
+      "learning_rate": 4.9963753072683025e-05,
+      "loss": 0.6164,
+      "step": 38
+    },
+    {
+      "epoch": 0.07046070460704607,
+      "grad_norm": 0.41115019726029983,
+      "learning_rate": 4.9955253244193375e-05,
+      "loss": 0.6182,
+      "step": 39
+    },
+    {
+      "epoch": 0.07226738934056007,
+      "grad_norm": 0.3495037257115665,
+      "learning_rate": 4.994585981845278e-05,
+      "loss": 0.6072,
+      "step": 40
+    },
+    {
+      "epoch": 0.07407407407407407,
+      "grad_norm": 0.3973831198418077,
+      "learning_rate": 4.9935573131820854e-05,
+      "loss": 0.6137,
+      "step": 41
+    },
+    {
+      "epoch": 0.07588075880758807,
+      "grad_norm": 0.3339390206679997,
+      "learning_rate": 4.9924393552643075e-05,
+      "loss": 0.6064,
+      "step": 42
+    },
+    {
+      "epoch": 0.07768744354110207,
+      "grad_norm": 0.3384678373992373,
+      "learning_rate": 4.991232148123761e-05,
+      "loss": 0.606,
+      "step": 43
+    },
+    {
+      "epoch": 0.07949412827461608,
+      "grad_norm": 0.3764492630812678,
+      "learning_rate": 4.989935734988098e-05,
+      "loss": 0.601,
+      "step": 44
+    },
+    {
+      "epoch": 0.08130081300813008,
+      "grad_norm": 0.35707353533847597,
+      "learning_rate": 4.988550162279255e-05,
+      "loss": 0.6097,
+      "step": 45
+    },
+    {
+      "epoch": 0.08310749774164408,
+      "grad_norm": 0.34040533355863656,
+      "learning_rate": 4.987075479611796e-05,
+      "loss": 0.6094,
+      "step": 46
+    },
+    {
+      "epoch": 0.08491418247515808,
+      "grad_norm": 0.34496537693497575,
+      "learning_rate": 4.985511739791129e-05,
+      "loss": 0.6129,
+      "step": 47
+    },
+    {
+      "epoch": 0.08672086720867209,
+      "grad_norm": 0.3655410266089117,
+      "learning_rate": 4.983858998811622e-05,
+      "loss": 0.5994,
+      "step": 48
+    },
+    {
+      "epoch": 0.08852755194218609,
+      "grad_norm": 0.32151319632227454,
+      "learning_rate": 4.9821173158545936e-05,
+      "loss": 0.607,
+      "step": 49
+    },
+    {
+      "epoch": 0.09033423667570009,
+      "grad_norm": 0.3573869377855791,
+      "learning_rate": 4.980286753286195e-05,
+      "loss": 0.6067,
+      "step": 50
+    },
+    {
+      "epoch": 0.0921409214092141,
+      "grad_norm": 0.3601371463623409,
+      "learning_rate": 4.978367376655177e-05,
+      "loss": 0.5965,
+      "step": 51
+    },
+    {
+      "epoch": 0.0939476061427281,
+      "grad_norm": 0.3288839599265164,
+      "learning_rate": 4.976359254690543e-05,
+      "loss": 0.6128,
+      "step": 52
+    },
+    {
+      "epoch": 0.0957542908762421,
+      "grad_norm": 0.3899144366212297,
+      "learning_rate": 4.974262459299087e-05,
+      "loss": 0.5992,
+      "step": 53
+    },
+    {
+      "epoch": 0.0975609756097561,
+      "grad_norm": 0.31373036051997816,
+      "learning_rate": 4.972077065562821e-05,
+      "loss": 0.5894,
+      "step": 54
+    },
+    {
+      "epoch": 0.0993676603432701,
+      "grad_norm": 0.3317064945312652,
+      "learning_rate": 4.969803151736284e-05,
+      "loss": 0.6046,
+      "step": 55
+    },
+    {
+      "epoch": 0.1011743450767841,
+      "grad_norm": 0.4008096818415782,
+      "learning_rate": 4.9674407992437394e-05,
+      "loss": 0.6091,
+      "step": 56
+    },
+    {
+      "epoch": 0.10298102981029811,
+      "grad_norm": 0.32205121061079506,
+      "learning_rate": 4.964990092676263e-05,
+      "loss": 0.5936,
+      "step": 57
+    },
+    {
+      "epoch": 0.10478771454381211,
+      "grad_norm": 0.3834592472356572,
+      "learning_rate": 4.962451119788709e-05,
+      "loss": 0.6035,
+      "step": 58
+    },
+    {
+      "epoch": 0.10659439927732611,
+      "grad_norm": 0.4364607834229955,
+      "learning_rate": 4.959823971496574e-05,
+      "loss": 0.5895,
+      "step": 59
+    },
+    {
+      "epoch": 0.10840108401084012,
+      "grad_norm": 0.39798614932188364,
+      "learning_rate": 4.957108741872736e-05,
+      "loss": 0.5903,
+      "step": 60
+    },
+    {
+      "epoch": 0.1102077687443541,
+      "grad_norm": 0.4078983073259993,
+      "learning_rate": 4.954305528144085e-05,
+      "loss": 0.594,
+      "step": 61
+    },
+    {
+      "epoch": 0.1120144534778681,
+      "grad_norm": 0.5064139886264023,
+      "learning_rate": 4.9514144306880506e-05,
+      "loss": 0.5989,
+      "step": 62
+    },
+    {
+      "epoch": 0.11382113821138211,
+      "grad_norm": 0.3049367414530056,
+      "learning_rate": 4.9484355530289944e-05,
+      "loss": 0.5982,
+      "step": 63
+    },
+    {
+      "epoch": 0.11562782294489611,
+      "grad_norm": 0.45932591541254997,
+      "learning_rate": 4.9453690018345144e-05,
+      "loss": 0.598,
+      "step": 64
+    },
+    {
+      "epoch": 0.11743450767841011,
+      "grad_norm": 0.34088884376991047,
+      "learning_rate": 4.9422148869116194e-05,
+      "loss": 0.5914,
+      "step": 65
+    },
+    {
+      "epoch": 0.11924119241192412,
+      "grad_norm": 0.4408196994719187,
+      "learning_rate": 4.938973321202799e-05,
+      "loss": 0.5943,
+      "step": 66
+    },
+    {
+      "epoch": 0.12104787714543812,
+      "grad_norm": 0.4034840944061305,
+      "learning_rate": 4.935644420781978e-05,
+      "loss": 0.5852,
+      "step": 67
+    },
+    {
+      "epoch": 0.12285456187895212,
+      "grad_norm": 0.3532258693927161,
+      "learning_rate": 4.932228304850363e-05,
+      "loss": 0.6003,
+      "step": 68
+    },
+    {
+      "epoch": 0.12466124661246612,
+      "grad_norm": 0.4041417285254443,
+      "learning_rate": 4.928725095732169e-05,
+      "loss": 0.6019,
+      "step": 69
+    },
+    {
+      "epoch": 0.12646793134598014,
+      "grad_norm": 0.338119605860217,
+      "learning_rate": 4.925134918870245e-05,
+      "loss": 0.6056,
+      "step": 70
+    },
+    {
+      "epoch": 0.12827461607949414,
+      "grad_norm": 0.2924641177682486,
+      "learning_rate": 4.9214579028215776e-05,
+      "loss": 0.5784,
+      "step": 71
+    },
+    {
+      "epoch": 0.13008130081300814,
+      "grad_norm": 0.37576892630911196,
+      "learning_rate": 4.917694179252692e-05,
+      "loss": 0.5966,
+      "step": 72
+    },
+    {
+      "epoch": 0.13188798554652212,
+      "grad_norm": 0.3066908816324021,
+      "learning_rate": 4.91384388293493e-05,
+      "loss": 0.5944,
+      "step": 73
+    },
+    {
+      "epoch": 0.13369467028003612,
+      "grad_norm": 0.3296017371410444,
+      "learning_rate": 4.909907151739633e-05,
+      "loss": 0.5863,
+      "step": 74
+    },
+    {
+      "epoch": 0.13550135501355012,
+      "grad_norm": 0.3510646842800697,
+      "learning_rate": 4.9058841266332e-05,
+      "loss": 0.5854,
+      "step": 75
+    },
+    {
+      "epoch": 0.13730803974706413,
+      "grad_norm": 0.33050013383675975,
+      "learning_rate": 4.90177495167204e-05,
+      "loss": 0.5816,
+      "step": 76
+    },
+    {
+      "epoch": 0.13911472448057813,
+      "grad_norm": 0.336545072709173,
+      "learning_rate": 4.897579773997415e-05,
+      "loss": 0.5768,
+      "step": 77
+    },
+    {
+      "epoch": 0.14092140921409213,
+      "grad_norm": 0.3401542906091868,
+      "learning_rate": 4.893298743830168e-05,
+      "loss": 0.5877,
+      "step": 78
+    },
+    {
+      "epoch": 0.14272809394760613,
+      "grad_norm": 0.30833272476392615,
+      "learning_rate": 4.888932014465352e-05,
+      "loss": 0.5949,
+      "step": 79
+    },
+    {
+      "epoch": 0.14453477868112014,
+      "grad_norm": 0.40829913126640544,
+      "learning_rate": 4.88447974226673e-05,
+      "loss": 0.6045,
+      "step": 80
+    },
+    {
+      "epoch": 0.14634146341463414,
+      "grad_norm": 0.3247095550760803,
+      "learning_rate": 4.879942086661184e-05,
+      "loss": 0.5901,
+      "step": 81
+    },
+    {
+      "epoch": 0.14814814814814814,
+      "grad_norm": 0.40484730371305205,
+      "learning_rate": 4.875319210133004e-05,
+      "loss": 0.5825,
+      "step": 82
+    },
+    {
+      "epoch": 0.14995483288166214,
+      "grad_norm": 0.37019522474324174,
+      "learning_rate": 4.870611278218066e-05,
+      "loss": 0.5918,
+      "step": 83
+    },
+    {
+      "epoch": 0.15176151761517614,
+      "grad_norm": 0.3654744041300334,
+      "learning_rate": 4.865818459497911e-05,
+      "loss": 0.5865,
+      "step": 84
+    },
+    {
+      "epoch": 0.15356820234869015,
+      "grad_norm": 0.3001653084612634,
+      "learning_rate": 4.860940925593703e-05,
+      "loss": 0.5889,
+      "step": 85
+    },
+    {
+      "epoch": 0.15537488708220415,
+      "grad_norm": 0.3463142769052332,
+      "learning_rate": 4.8559788511600876e-05,
+      "loss": 0.5881,
+      "step": 86
+    },
+    {
+      "epoch": 0.15718157181571815,
+      "grad_norm": 0.3221292284344934,
+      "learning_rate": 4.850932413878934e-05,
+      "loss": 0.5901,
+      "step": 87
+    },
+    {
+      "epoch": 0.15898825654923215,
+      "grad_norm": 0.33253415770699135,
+      "learning_rate": 4.8458017944529776e-05,
+      "loss": 0.5952,
+      "step": 88
+    },
+    {
+      "epoch": 0.16079494128274616,
+      "grad_norm": 0.37021022779245716,
+      "learning_rate": 4.8405871765993433e-05,
+      "loss": 0.5928,
+      "step": 89
+    },
+    {
+      "epoch": 0.16260162601626016,
+      "grad_norm": 0.3197338147243217,
+      "learning_rate": 4.8352887470429726e-05,
+      "loss": 0.5837,
+      "step": 90
+    },
+    {
+      "epoch": 0.16440831074977416,
+      "grad_norm": 0.35706780201968874,
+      "learning_rate": 4.8299066955099335e-05,
+      "loss": 0.5811,
+      "step": 91
+    },
+    {
+      "epoch": 0.16621499548328816,
+      "grad_norm": 0.3730971582648582,
+      "learning_rate": 4.8244412147206284e-05,
+      "loss": 0.586,
+      "step": 92
+    },
+    {
+      "epoch": 0.16802168021680217,
+      "grad_norm": 0.33467520904104064,
+      "learning_rate": 4.8188925003828945e-05,
+      "loss": 0.5919,
+      "step": 93
+    },
+    {
+      "epoch": 0.16982836495031617,
+      "grad_norm": 0.41181901183464464,
+      "learning_rate": 4.813260751184992e-05,
+      "loss": 0.5922,
+      "step": 94
+    },
+    {
+      "epoch": 0.17163504968383017,
+      "grad_norm": 0.31057237794043846,
+      "learning_rate": 4.807546168788494e-05,
+      "loss": 0.5835,
+      "step": 95
+    },
+    {
+      "epoch": 0.17344173441734417,
+      "grad_norm": 0.4045185112342142,
+      "learning_rate": 4.8017489578210604e-05,
+      "loss": 0.5839,
+      "step": 96
+    },
+    {
+      "epoch": 0.17524841915085818,
+      "grad_norm": 0.342091450071029,
+      "learning_rate": 4.7958693258691167e-05,
+      "loss": 0.5891,
+      "step": 97
+    },
+    {
+      "epoch": 0.17705510388437218,
+      "grad_norm": 0.32193383766669476,
+      "learning_rate": 4.7899074834704165e-05,
+      "loss": 0.5814,
+      "step": 98
+    },
+    {
+      "epoch": 0.17886178861788618,
+      "grad_norm": 0.33881983844597735,
+      "learning_rate": 4.783863644106502e-05,
+      "loss": 0.5951,
+      "step": 99
+    },
+    {
+      "epoch": 0.18066847335140018,
+      "grad_norm": 0.3141049036238513,
+      "learning_rate": 4.7777380241950645e-05,
+      "loss": 0.5672,
+      "step": 100
+    },
+    {
+      "epoch": 0.18247515808491419,
+      "grad_norm": 0.30258609960076854,
+      "learning_rate": 4.7715308430821864e-05,
+      "loss": 0.5831,
+      "step": 101
+    },
+    {
+      "epoch": 0.1842818428184282,
+      "grad_norm": 0.33013650150122287,
+      "learning_rate": 4.765242323034498e-05,
+      "loss": 0.6019,
+      "step": 102
+    },
+    {
+      "epoch": 0.1860885275519422,
+      "grad_norm": 0.31075076006352215,
+      "learning_rate": 4.758872689231208e-05,
+      "loss": 0.581,
+      "step": 103
+    },
+    {
+      "epoch": 0.1878952122854562,
+      "grad_norm": 0.3376659425636789,
+      "learning_rate": 4.752422169756048e-05,
+      "loss": 0.5775,
+      "step": 104
+    },
+    {
+      "epoch": 0.1897018970189702,
+      "grad_norm": 0.3373397970824128,
+      "learning_rate": 4.745890995589101e-05,
+      "loss": 0.5705,
+      "step": 105
+    },
+    {
+      "epoch": 0.1915085817524842,
+      "grad_norm": 0.3159036744458245,
+      "learning_rate": 4.7392794005985326e-05,
+      "loss": 0.574,
+      "step": 106
+    },
+    {
+      "epoch": 0.1933152664859982,
+      "grad_norm": 0.3157369760717556,
+      "learning_rate": 4.732587621532214e-05,
+      "loss": 0.58,
+      "step": 107
+    },
+    {
+      "epoch": 0.1951219512195122,
+      "grad_norm": 0.3151534184596026,
+      "learning_rate": 4.725815898009247e-05,
+      "loss": 0.5785,
+      "step": 108
+    },
+    {
+      "epoch": 0.1969286359530262,
+      "grad_norm": 0.3067792635656275,
+      "learning_rate": 4.718964472511386e-05,
+      "loss": 0.5864,
+      "step": 109
+    },
+    {
+      "epoch": 0.1987353206865402,
+      "grad_norm": 0.34885189836356073,
+      "learning_rate": 4.712033590374346e-05,
+      "loss": 0.5685,
+      "step": 110
+    },
+    {
+      "epoch": 0.2005420054200542,
+      "grad_norm": 0.2725346961548414,
+      "learning_rate": 4.705023499779031e-05,
+      "loss": 0.5876,
+      "step": 111
+    },
+    {
+      "epoch": 0.2023486901535682,
+      "grad_norm": 0.34146491162388654,
+      "learning_rate": 4.6979344517426345e-05,
+      "loss": 0.5814,
+      "step": 112
+    },
+    {
+      "epoch": 0.2041553748870822,
+      "grad_norm": 0.2979010304617956,
+      "learning_rate": 4.690766700109659e-05,
+      "loss": 0.5857,
+      "step": 113
+    },
+    {
+      "epoch": 0.20596205962059622,
+      "grad_norm": 0.3317475544222318,
+      "learning_rate": 4.6835205015428246e-05,
+      "loss": 0.5685,
+      "step": 114
+    },
+    {
+      "epoch": 0.20776874435411022,
+      "grad_norm": 0.3490432374306535,
+      "learning_rate": 4.676196115513876e-05,
+      "loss": 0.585,
+      "step": 115
+    },
+    {
+      "epoch": 0.20957542908762422,
+      "grad_norm": 0.2845519792036694,
+      "learning_rate": 4.668793804294294e-05,
+      "loss": 0.5779,
+      "step": 116
+    },
+    {
+      "epoch": 0.21138211382113822,
+      "grad_norm": 0.30626266068728963,
+      "learning_rate": 4.661313832945904e-05,
+      "loss": 0.5901,
+      "step": 117
+    },
+    {
+      "epoch": 0.21318879855465223,
+      "grad_norm": 0.3411792829442568,
+      "learning_rate": 4.653756469311381e-05,
+      "loss": 0.5766,
+      "step": 118
+    },
+    {
+      "epoch": 0.21499548328816623,
+      "grad_norm": 0.273095998341187,
+      "learning_rate": 4.6461219840046654e-05,
+      "loss": 0.5666,
+      "step": 119
+    },
+    {
+      "epoch": 0.21680216802168023,
+      "grad_norm": 0.3070242568215688,
+      "learning_rate": 4.638410650401267e-05,
+      "loss": 0.5861,
+      "step": 120
+    },
+    {
+      "epoch": 0.2186088527551942,
+      "grad_norm": 0.315496788680795,
+      "learning_rate": 4.6306227446284775e-05,
+      "loss": 0.5817,
+      "step": 121
+    },
+    {
+      "epoch": 0.2204155374887082,
+      "grad_norm": 0.285486208489652,
+      "learning_rate": 4.622758545555485e-05,
+      "loss": 0.5862,
+      "step": 122
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.3014286983026253,
+      "learning_rate": 4.614818334783384e-05,
+      "loss": 0.5756,
+      "step": 123
+    },
+    {
+      "epoch": 0.2240289069557362,
+      "grad_norm": 0.3286311142757772,
+      "learning_rate": 4.606802396635098e-05,
+      "loss": 0.5622,
+      "step": 124
+    },
+    {
+      "epoch": 0.22583559168925021,
+      "grad_norm": 0.3099894773692972,
+      "learning_rate": 4.598711018145193e-05,
+      "loss": 0.5702,
+      "step": 125
+    },
+    {
+      "epoch": 0.22764227642276422,
+      "grad_norm": 0.3402171436347276,
+      "learning_rate": 4.590544489049602e-05,
+      "loss": 0.5822,
+      "step": 126
+    },
+    {
+      "epoch": 0.22944896115627822,
+      "grad_norm": 0.2949342532195096,
+      "learning_rate": 4.5823031017752485e-05,
+      "loss": 0.5753,
+      "step": 127
+    },
+    {
+      "epoch": 0.23125564588979222,
+      "grad_norm": 0.33271367144515557,
+      "learning_rate": 4.5739871514295786e-05,
+      "loss": 0.5746,
+      "step": 128
+    },
+    {
+      "epoch": 0.23306233062330622,
+      "grad_norm": 0.2854287194746108,
+      "learning_rate": 4.5655969357899874e-05,
+      "loss": 0.5688,
+      "step": 129
+    },
+    {
+      "epoch": 0.23486901535682023,
+      "grad_norm": 0.2911634608087971,
+      "learning_rate": 4.5571327552931645e-05,
+      "loss": 0.5875,
+      "step": 130
+    },
+    {
+      "epoch": 0.23667570009033423,
+      "grad_norm": 0.3282856212432573,
+      "learning_rate": 4.54859491302433e-05,
+      "loss": 0.5828,
+      "step": 131
+    },
+    {
+      "epoch": 0.23848238482384823,
+      "grad_norm": 0.31230217501155305,
+      "learning_rate": 4.5399837147063825e-05,
+      "loss": 0.5622,
+      "step": 132
+    },
+    {
+      "epoch": 0.24028906955736223,
+      "grad_norm": 0.26497408669368616,
+      "learning_rate": 4.531299468688955e-05,
+      "loss": 0.576,
+      "step": 133
+    },
+    {
+      "epoch": 0.24209575429087624,
+      "grad_norm": 0.31735124152798094,
+      "learning_rate": 4.522542485937369e-05,
+      "loss": 0.5772,
+      "step": 134
+    },
+    {
+      "epoch": 0.24390243902439024,
+      "grad_norm": 0.3015639590739915,
+      "learning_rate": 4.5137130800215025e-05,
+      "loss": 0.5748,
+      "step": 135
+    },
+    {
+      "epoch": 0.24570912375790424,
+      "grad_norm": 0.27756271594362325,
+      "learning_rate": 4.50481156710456e-05,
+      "loss": 0.5782,
+      "step": 136
+    },
+    {
+      "epoch": 0.24751580849141824,
+      "grad_norm": 0.3311373615406198,
+      "learning_rate": 4.495838265931754e-05,
+      "loss": 0.5632,
+      "step": 137
+    },
+    {
+      "epoch": 0.24932249322493225,
+      "grad_norm": 0.26966897365996273,
+      "learning_rate": 4.486793497818889e-05,
+      "loss": 0.5778,
+      "step": 138
+    },
+    {
+      "epoch": 0.25112917795844625,
+      "grad_norm": 0.3458179405656153,
+      "learning_rate": 4.477677586640854e-05,
+      "loss": 0.5814,
+      "step": 139
+    },
+    {
+      "epoch": 0.2529358626919603,
+      "grad_norm": 0.2722125808652539,
+      "learning_rate": 4.4684908588200304e-05,
+      "loss": 0.5804,
+      "step": 140
+    },
+    {
+      "epoch": 0.25474254742547425,
+      "grad_norm": 0.29853453900528515,
+      "learning_rate": 4.4592336433146e-05,
+      "loss": 0.5622,
+      "step": 141
+    },
+    {
+      "epoch": 0.2565492321589883,
+      "grad_norm": 0.2616025861086357,
+      "learning_rate": 4.449906271606766e-05,
+      "loss": 0.5816,
+      "step": 142
+    },
+    {
+      "epoch": 0.25835591689250226,
+      "grad_norm": 0.28875897468487793,
+      "learning_rate": 4.440509077690883e-05,
+      "loss": 0.5751,
+      "step": 143
+    },
+    {
+      "epoch": 0.2601626016260163,
+      "grad_norm": 0.31089280557100357,
+      "learning_rate": 4.431042398061499e-05,
+      "loss": 0.5614,
+      "step": 144
+    },
+    {
+      "epoch": 0.26196928635953026,
+      "grad_norm": 0.24453740569318233,
+      "learning_rate": 4.421506571701305e-05,
+      "loss": 0.5739,
+      "step": 145
+    },
+    {
+      "epoch": 0.26377597109304424,
+      "grad_norm": 0.3298892559409927,
+      "learning_rate": 4.4119019400689967e-05,
+      "loss": 0.5789,
+      "step": 146
+    },
+    {
+      "epoch": 0.26558265582655827,
+      "grad_norm": 0.29175370949614915,
+      "learning_rate": 4.402228847087047e-05,
+      "loss": 0.5687,
+      "step": 147
+    },
+    {
+      "epoch": 0.26738934056007224,
+      "grad_norm": 0.30364879113096566,
+      "learning_rate": 4.3924876391293915e-05,
+      "loss": 0.57,
+      "step": 148
+    },
+    {
+      "epoch": 0.26919602529358627,
+      "grad_norm": 0.2924856865957413,
+      "learning_rate": 4.382678665009028e-05,
+      "loss": 0.5675,
+      "step": 149
+    },
+    {
+      "epoch": 0.27100271002710025,
+      "grad_norm": 0.2831086406138475,
+      "learning_rate": 4.372802275965521e-05,
+      "loss": 0.5777,
+      "step": 150
+    },
+    {
+      "epoch": 0.2728093947606143,
+      "grad_norm": 0.3017093604784243,
+      "learning_rate": 4.3628588256524285e-05,
+      "loss": 0.5589,
+      "step": 151
+    },
+    {
+      "epoch": 0.27461607949412825,
+      "grad_norm": 0.2765860953865558,
+      "learning_rate": 4.3528486701246376e-05,
+      "loss": 0.5669,
+      "step": 152
+    },
+    {
+      "epoch": 0.2764227642276423,
+      "grad_norm": 0.2904031273845622,
+      "learning_rate": 4.3427721678256125e-05,
+      "loss": 0.5671,
+      "step": 153
+    },
+    {
+      "epoch": 0.27822944896115626,
+      "grad_norm": 0.2540349949087337,
+      "learning_rate": 4.332629679574566e-05,
+      "loss": 0.5702,
+      "step": 154
+    },
+    {
+      "epoch": 0.2800361336946703,
+      "grad_norm": 0.3069633678570879,
+      "learning_rate": 4.3224215685535294e-05,
+      "loss": 0.5637,
+      "step": 155
+    },
+    {
+      "epoch": 0.28184281842818426,
+      "grad_norm": 0.2542441755605158,
+      "learning_rate": 4.312148200294355e-05,
+      "loss": 0.5627,
+      "step": 156
+    },
+    {
+      "epoch": 0.2836495031616983,
+      "grad_norm": 0.3018408188777611,
+      "learning_rate": 4.301809942665625e-05,
+      "loss": 0.5757,
+      "step": 157
+    },
+    {
+      "epoch": 0.28545618789521227,
+      "grad_norm": 0.28743959410933695,
+      "learning_rate": 4.2914071658594805e-05,
+      "loss": 0.5734,
+      "step": 158
+    },
+    {
+      "epoch": 0.2872628726287263,
+      "grad_norm": 0.2787808725595535,
+      "learning_rate": 4.2809402423783624e-05,
+      "loss": 0.5542,
+      "step": 159
+    },
+    {
+      "epoch": 0.28906955736224027,
+      "grad_norm": 0.29206183996226404,
+      "learning_rate": 4.2704095470216744e-05,
+      "loss": 0.5635,
+      "step": 160
+    },
+    {
+      "epoch": 0.2908762420957543,
+      "grad_norm": 0.28832091136901655,
+      "learning_rate": 4.2598154568723626e-05,
+      "loss": 0.5716,
+      "step": 161
+    },
+    {
+      "epoch": 0.2926829268292683,
+      "grad_norm": 0.26053870722630096,
+      "learning_rate": 4.249158351283414e-05,
+      "loss": 0.5568,
+      "step": 162
+    },
+    {
+      "epoch": 0.2944896115627823,
+      "grad_norm": 0.30602256840625275,
+      "learning_rate": 4.2384386118642694e-05,
+      "loss": 0.5661,
+      "step": 163
+    },
+    {
+      "epoch": 0.2962962962962963,
+      "grad_norm": 0.24580296790912456,
+      "learning_rate": 4.227656622467162e-05,
+      "loss": 0.5751,
+      "step": 164
+    },
+    {
+      "epoch": 0.2981029810298103,
+      "grad_norm": 0.29059059202913196,
+      "learning_rate": 4.2168127691733706e-05,
+      "loss": 0.5662,
+      "step": 165
+    },
+    {
+      "epoch": 0.2999096657633243,
+      "grad_norm": 0.29414515714153977,
+      "learning_rate": 4.205907440279395e-05,
+      "loss": 0.5724,
+      "step": 166
+    },
+    {
+      "epoch": 0.3017163504968383,
+      "grad_norm": 0.29414614437874254,
+      "learning_rate": 4.1949410262830525e-05,
+      "loss": 0.5717,
+      "step": 167
+    },
+    {
+      "epoch": 0.3035230352303523,
+      "grad_norm": 0.2971349224474148,
+      "learning_rate": 4.1839139198694946e-05,
+      "loss": 0.5697,
+      "step": 168
+    },
+    {
+      "epoch": 0.3053297199638663,
+      "grad_norm": 0.2793813743979144,
+      "learning_rate": 4.172826515897146e-05,
+      "loss": 0.5617,
+      "step": 169
+    },
+    {
+      "epoch": 0.3071364046973803,
+      "grad_norm": 0.3023179135068114,
+      "learning_rate": 4.161679211383565e-05,
+      "loss": 0.5751,
+      "step": 170
+    },
+    {
+      "epoch": 0.3089430894308943,
+      "grad_norm": 0.2477222878497878,
+      "learning_rate": 4.150472405491226e-05,
+      "loss": 0.5583,
+      "step": 171
+    },
+    {
+      "epoch": 0.3107497741644083,
+      "grad_norm": 0.2707854956368433,
+      "learning_rate": 4.139206499513231e-05,
+      "loss": 0.5658,
+      "step": 172
+    },
+    {
+      "epoch": 0.31255645889792233,
+      "grad_norm": 0.26805394435378294,
+      "learning_rate": 4.127881896858934e-05,
+      "loss": 0.5687,
+      "step": 173
+    },
+    {
+      "epoch": 0.3143631436314363,
+      "grad_norm": 0.24818328664431513,
+      "learning_rate": 4.116499003039499e-05,
+      "loss": 0.5597,
+      "step": 174
+    },
+    {
+      "epoch": 0.31616982836495033,
+      "grad_norm": 0.3076486841492656,
+      "learning_rate": 4.105058225653381e-05,
+      "loss": 0.5638,
+      "step": 175
+    },
+    {
+      "epoch": 0.3179765130984643,
+      "grad_norm": 0.24484381551700804,
+      "learning_rate": 4.093559974371725e-05,
+      "loss": 0.578,
+      "step": 176
+    },
+    {
+      "epoch": 0.31978319783197834,
+      "grad_norm": 0.25779831771912026,
+      "learning_rate": 4.082004660923703e-05,
+      "loss": 0.5536,
+      "step": 177
+    },
+    {
+      "epoch": 0.3215898825654923,
+      "grad_norm": 0.2431924544197558,
+      "learning_rate": 4.070392699081767e-05,
+      "loss": 0.5606,
+      "step": 178
+    },
+    {
+      "epoch": 0.32339656729900634,
+      "grad_norm": 0.24360453043613553,
+      "learning_rate": 4.058724504646834e-05,
+      "loss": 0.571,
+      "step": 179
+    },
+    {
+      "epoch": 0.3252032520325203,
+      "grad_norm": 0.28406098303419186,
+      "learning_rate": 4.047000495433397e-05,
+      "loss": 0.5781,
+      "step": 180
+    },
+    {
+      "epoch": 0.32700993676603435,
+      "grad_norm": 0.22058075555118684,
+      "learning_rate": 4.035221091254563e-05,
+      "loss": 0.5574,
+      "step": 181
+    },
+    {
+      "epoch": 0.3288166214995483,
+      "grad_norm": 0.3009339432220675,
+      "learning_rate": 4.023386713907021e-05,
+      "loss": 0.5677,
+      "step": 182
+    },
+    {
+      "epoch": 0.33062330623306235,
+      "grad_norm": 0.26774508776974565,
+      "learning_rate": 4.011497787155938e-05,
+      "loss": 0.577,
+      "step": 183
+    },
+    {
+      "epoch": 0.3324299909665763,
+      "grad_norm": 0.284303170530224,
+      "learning_rate": 3.9995547367197845e-05,
+      "loss": 0.5735,
+      "step": 184
+    },
+    {
+      "epoch": 0.33423667570009036,
+      "grad_norm": 0.2614285883935899,
+      "learning_rate": 3.987557990255093e-05,
+      "loss": 0.5695,
+      "step": 185
+    },
+    {
+      "epoch": 0.33604336043360433,
+      "grad_norm": 0.28256380914027235,
+      "learning_rate": 3.975507977341141e-05,
+      "loss": 0.5648,
+      "step": 186
+    },
+    {
+      "epoch": 0.33785004516711836,
+      "grad_norm": 0.28945389202706384,
+      "learning_rate": 3.963405129464569e-05,
+      "loss": 0.5618,
+      "step": 187
+    },
+    {
+      "epoch": 0.33965672990063234,
+      "grad_norm": 0.2654043705467122,
+      "learning_rate": 3.9512498800039335e-05,
+      "loss": 0.5622,
+      "step": 188
+    },
+    {
+      "epoch": 0.34146341463414637,
+      "grad_norm": 0.28212350075401693,
+      "learning_rate": 3.939042664214184e-05,
+      "loss": 0.5691,
+      "step": 189
+    },
+    {
+      "epoch": 0.34327009936766034,
+      "grad_norm": 0.2666801982334387,
+      "learning_rate": 3.92678391921108e-05,
+      "loss": 0.5582,
+      "step": 190
+    },
+    {
+      "epoch": 0.34507678410117437,
+      "grad_norm": 0.2658380267209616,
+      "learning_rate": 3.914474083955537e-05,
+      "loss": 0.5561,
+      "step": 191
+    },
+    {
+      "epoch": 0.34688346883468835,
+      "grad_norm": 0.2767810249999182,
+      "learning_rate": 3.902113599237911e-05,
+      "loss": 0.5637,
+      "step": 192
+    },
+    {
+      "epoch": 0.3486901535682023,
+      "grad_norm": 0.2686248952050933,
+      "learning_rate": 3.8897029076622116e-05,
+      "loss": 0.5648,
+      "step": 193
+    },
+    {
+      "epoch": 0.35049683830171635,
+      "grad_norm": 0.2671044597386809,
+      "learning_rate": 3.8772424536302564e-05,
+      "loss": 0.5668,
+      "step": 194
+    },
+    {
+      "epoch": 0.3523035230352303,
+      "grad_norm": 0.28240978604941985,
+      "learning_rate": 3.8647326833257545e-05,
+      "loss": 0.5503,
+      "step": 195
+    },
+    {
+      "epoch": 0.35411020776874436,
+      "grad_norm": 0.2545745653651051,
+      "learning_rate": 3.852174044698333e-05,
+      "loss": 0.564,
+      "step": 196
+    },
+    {
+      "epoch": 0.35591689250225833,
+      "grad_norm": 0.2788914079559153,
+      "learning_rate": 3.8395669874474915e-05,
+      "loss": 0.563,
+      "step": 197
+    },
+    {
+      "epoch": 0.35772357723577236,
+      "grad_norm": 0.27664913837367605,
+      "learning_rate": 3.826911963006507e-05,
+      "loss": 0.5397,
+      "step": 198
+    },
+    {
+      "epoch": 0.35953026196928634,
+      "grad_norm": 0.2854592929422416,
+      "learning_rate": 3.814209424526262e-05,
+      "loss": 0.5646,
+      "step": 199
+    },
+    {
+      "epoch": 0.36133694670280037,
+      "grad_norm": 0.31461524287063203,
+      "learning_rate": 3.801459826859022e-05,
+      "loss": 0.5521,
+      "step": 200
+    },
+    {
+      "epoch": 0.36314363143631434,
+      "grad_norm": 0.2571618033810251,
+      "learning_rate": 3.788663626542146e-05,
+      "loss": 0.547,
+      "step": 201
+    },
+    {
+      "epoch": 0.36495031616982837,
+      "grad_norm": 0.2698316727635178,
+      "learning_rate": 3.7758212817817405e-05,
+      "loss": 0.5735,
+      "step": 202
+    },
+    {
+      "epoch": 0.36675700090334235,
+      "grad_norm": 0.28386625197611576,
+      "learning_rate": 3.762933252436253e-05,
+      "loss": 0.5607,
+      "step": 203
+    },
+    {
+      "epoch": 0.3685636856368564,
+      "grad_norm": 0.23910352580001357,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.5729,
+      "step": 204
+    },
+    {
+      "epoch": 0.37037037037037035,
+      "grad_norm": 0.25373354611118293,
+      "learning_rate": 3.73702198758665e-05,
+      "loss": 0.5636,
+      "step": 205
+    },
+    {
+      "epoch": 0.3721770551038844,
+      "grad_norm": 0.26220155901200526,
+      "learning_rate": 3.7239996799126314e-05,
+      "loss": 0.5511,
+      "step": 206
+    },
+    {
+      "epoch": 0.37398373983739835,
+      "grad_norm": 0.2703230132182625,
+      "learning_rate": 3.7109335432805006e-05,
+      "loss": 0.559,
+      "step": 207
+    },
+    {
+      "epoch": 0.3757904245709124,
+      "grad_norm": 0.25300776840046674,
+      "learning_rate": 3.697824045562238e-05,
+      "loss": 0.5467,
+      "step": 208
+    },
+    {
+      "epoch": 0.37759710930442636,
+      "grad_norm": 0.31024727576581884,
+      "learning_rate": 3.6846716561824965e-05,
+      "loss": 0.5681,
+      "step": 209
+    },
+    {
+      "epoch": 0.3794037940379404,
+      "grad_norm": 0.2547242302465973,
+      "learning_rate": 3.6714768461017965e-05,
+      "loss": 0.5641,
+      "step": 210
+    },
+    {
+      "epoch": 0.38121047877145436,
+      "grad_norm": 0.2832537591345291,
+      "learning_rate": 3.6582400877996546e-05,
+      "loss": 0.5631,
+      "step": 211
+    },
+    {
+      "epoch": 0.3830171635049684,
+      "grad_norm": 0.23740735296590512,
+      "learning_rate": 3.6449618552576695e-05,
+      "loss": 0.5705,
+      "step": 212
+    },
+    {
+      "epoch": 0.38482384823848237,
+      "grad_norm": 0.25138099601672315,
+      "learning_rate": 3.6316426239425485e-05,
+      "loss": 0.5712,
+      "step": 213
+    },
+    {
+      "epoch": 0.3866305329719964,
+      "grad_norm": 0.2446936529822529,
+      "learning_rate": 3.6182828707890816e-05,
+      "loss": 0.5574,
+      "step": 214
+    },
+    {
+      "epoch": 0.3884372177055104,
+      "grad_norm": 0.28278554244015486,
+      "learning_rate": 3.604883074183068e-05,
+      "loss": 0.5848,
+      "step": 215
+    },
+    {
+      "epoch": 0.3902439024390244,
+      "grad_norm": 0.2868056773879424,
+      "learning_rate": 3.591443713944175e-05,
+      "loss": 0.5728,
+      "step": 216
+    },
+    {
+      "epoch": 0.3920505871725384,
+      "grad_norm": 0.23733474164783092,
+      "learning_rate": 3.577965271308771e-05,
+      "loss": 0.5734,
+      "step": 217
+    },
+    {
+      "epoch": 0.3938572719060524,
+      "grad_norm": 0.2418348363911958,
+      "learning_rate": 3.564448228912682e-05,
+      "loss": 0.5644,
+      "step": 218
+    },
+    {
+      "epoch": 0.3956639566395664,
+      "grad_norm": 0.24258866663076342,
+      "learning_rate": 3.550893070773914e-05,
+      "loss": 0.5597,
+      "step": 219
+    },
+    {
+      "epoch": 0.3974706413730804,
+      "grad_norm": 0.23248379507210618,
+      "learning_rate": 3.5373002822753216e-05,
+      "loss": 0.5619,
+      "step": 220
+    },
+    {
+      "epoch": 0.3992773261065944,
+      "grad_norm": 0.23221387790837575,
+      "learning_rate": 3.5236703501472266e-05,
+      "loss": 0.551,
+      "step": 221
+    },
+    {
+      "epoch": 0.4010840108401084,
+      "grad_norm": 0.27427489494087726,
+      "learning_rate": 3.510003762449988e-05,
+      "loss": 0.5685,
+      "step": 222
+    },
+    {
+      "epoch": 0.4028906955736224,
+      "grad_norm": 0.25685297943543806,
+      "learning_rate": 3.496301008556529e-05,
+      "loss": 0.5508,
+      "step": 223
+    },
+    {
+      "epoch": 0.4046973803071364,
+      "grad_norm": 0.23751224178133434,
+      "learning_rate": 3.4825625791348096e-05,
+      "loss": 0.5553,
+      "step": 224
+    },
+    {
+      "epoch": 0.4065040650406504,
+      "grad_norm": 0.2533622057517779,
+      "learning_rate": 3.4687889661302576e-05,
+      "loss": 0.5616,
+      "step": 225
+    },
+    {
+      "epoch": 0.4083107497741644,
+      "grad_norm": 0.25908397166669356,
+      "learning_rate": 3.454980662748156e-05,
+      "loss": 0.5659,
+      "step": 226
+    },
+    {
+      "epoch": 0.4101174345076784,
+      "grad_norm": 0.25053206427350594,
+      "learning_rate": 3.44113816343598e-05,
+      "loss": 0.5701,
+      "step": 227
+    },
+    {
+      "epoch": 0.41192411924119243,
+      "grad_norm": 0.2371124422311737,
+      "learning_rate": 3.427261963865691e-05,
+      "loss": 0.5567,
+      "step": 228
+    },
+    {
+      "epoch": 0.4137308039747064,
+      "grad_norm": 0.26537475430824775,
+      "learning_rate": 3.413352560915988e-05,
+      "loss": 0.5594,
+      "step": 229
+    },
+    {
+      "epoch": 0.41553748870822044,
+      "grad_norm": 0.24213575293916792,
+      "learning_rate": 3.399410452654518e-05,
+      "loss": 0.5612,
+      "step": 230
+    },
+    {
+      "epoch": 0.4173441734417344,
+      "grad_norm": 0.24275309716779916,
+      "learning_rate": 3.3854361383200374e-05,
+      "loss": 0.561,
+      "step": 231
+    },
+    {
+      "epoch": 0.41915085817524844,
+      "grad_norm": 0.23843095168601489,
+      "learning_rate": 3.3714301183045385e-05,
+      "loss": 0.5631,
+      "step": 232
+    },
+    {
+      "epoch": 0.4209575429087624,
+      "grad_norm": 0.21791224002383097,
+      "learning_rate": 3.357392894135329e-05,
+      "loss": 0.5457,
+      "step": 233
+    },
+    {
+      "epoch": 0.42276422764227645,
+      "grad_norm": 0.23785549929021552,
+      "learning_rate": 3.343324968457076e-05,
+      "loss": 0.5608,
+      "step": 234
+    },
+    {
+      "epoch": 0.4245709123757904,
+      "grad_norm": 0.22396755626120254,
+      "learning_rate": 3.329226845013802e-05,
+      "loss": 0.5558,
+      "step": 235
+    },
+    {
+      "epoch": 0.42637759710930445,
+      "grad_norm": 0.22217117995010718,
+      "learning_rate": 3.315099028630855e-05,
+      "loss": 0.5669,
+      "step": 236
+    },
+    {
+      "epoch": 0.4281842818428184,
+      "grad_norm": 0.2368068176727353,
+      "learning_rate": 3.3009420251968244e-05,
+      "loss": 0.5676,
+      "step": 237
+    },
+    {
+      "epoch": 0.42999096657633246,
+      "grad_norm": 0.2913166749962225,
+      "learning_rate": 3.28675634164543e-05,
+      "loss": 0.5757,
+      "step": 238
+    },
+    {
+      "epoch": 0.43179765130984643,
+      "grad_norm": 0.2326443530840128,
+      "learning_rate": 3.272542485937369e-05,
+      "loss": 0.5562,
+      "step": 239
+    },
+    {
+      "epoch": 0.43360433604336046,
+      "grad_norm": 0.21709853787458164,
+      "learning_rate": 3.258300967042125e-05,
+      "loss": 0.5566,
+      "step": 240
+    },
+    {
+      "epoch": 0.43541102077687444,
+      "grad_norm": 0.24943018423195573,
+      "learning_rate": 3.244032294919747e-05,
+      "loss": 0.5558,
+      "step": 241
+    },
+    {
+      "epoch": 0.4372177055103884,
+      "grad_norm": 0.2088255652107171,
+      "learning_rate": 3.229736980502584e-05,
+      "loss": 0.5561,
+      "step": 242
+    },
+    {
+      "epoch": 0.43902439024390244,
+      "grad_norm": 0.2753534476789369,
+      "learning_rate": 3.215415535676992e-05,
+      "loss": 0.561,
+      "step": 243
+    },
+    {
+      "epoch": 0.4408310749774164,
+      "grad_norm": 0.21949470480639668,
+      "learning_rate": 3.201068473265007e-05,
+      "loss": 0.5681,
+      "step": 244
+    },
+    {
+      "epoch": 0.44263775971093045,
+      "grad_norm": 0.26944393055258614,
+      "learning_rate": 3.186696307005976e-05,
+      "loss": 0.5591,
+      "step": 245
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.2191119214759813,
+      "learning_rate": 3.172299551538164e-05,
+      "loss": 0.566,
+      "step": 246
+    },
+    {
+      "epoch": 0.44625112917795845,
+      "grad_norm": 0.23598250909414767,
+      "learning_rate": 3.15787872238033e-05,
+      "loss": 0.5474,
+      "step": 247
+    },
+    {
+      "epoch": 0.4480578139114724,
+      "grad_norm": 0.2244716812824527,
+      "learning_rate": 3.143434335913256e-05,
+      "loss": 0.5665,
+      "step": 248
+    },
+    {
+      "epoch": 0.44986449864498645,
+      "grad_norm": 0.2605217389849008,
+      "learning_rate": 3.1289669093612714e-05,
+      "loss": 0.55,
+      "step": 249
+    },
+    {
+      "epoch": 0.45167118337850043,
+      "grad_norm": 0.23565788514397004,
+      "learning_rate": 3.1144769607737204e-05,
+      "loss": 0.5648,
+      "step": 250
+    },
+    {
+      "epoch": 0.45347786811201446,
+      "grad_norm": 0.2520253908292615,
+      "learning_rate": 3.099965009006415e-05,
+      "loss": 0.5515,
+      "step": 251
+    },
+    {
+      "epoch": 0.45528455284552843,
+      "grad_norm": 0.23382294482671243,
+      "learning_rate": 3.0854315737030596e-05,
+      "loss": 0.5343,
+      "step": 252
+    },
+    {
+      "epoch": 0.45709123757904246,
+      "grad_norm": 0.21156862362731088,
+      "learning_rate": 3.0708771752766394e-05,
+      "loss": 0.5566,
+      "step": 253
+    },
+    {
+      "epoch": 0.45889792231255644,
+      "grad_norm": 0.2858583262540506,
+      "learning_rate": 3.056302334890786e-05,
+      "loss": 0.5562,
+      "step": 254
+    },
+    {
+      "epoch": 0.46070460704607047,
+      "grad_norm": 0.21446015297714893,
+      "learning_rate": 3.0417075744411178e-05,
+      "loss": 0.5491,
+      "step": 255
+    },
+    {
+      "epoch": 0.46251129177958444,
+      "grad_norm": 0.27175171429400885,
+      "learning_rate": 3.0270934165365478e-05,
+      "loss": 0.549,
+      "step": 256
+    },
+    {
+      "epoch": 0.4643179765130985,
+      "grad_norm": 0.2332022756162156,
+      "learning_rate": 3.0124603844805767e-05,
+      "loss": 0.5615,
+      "step": 257
+    },
+    {
+      "epoch": 0.46612466124661245,
+      "grad_norm": 0.25726612906200885,
+      "learning_rate": 2.997809002252546e-05,
+      "loss": 0.5545,
+      "step": 258
+    },
+    {
+      "epoch": 0.4679313459801265,
+      "grad_norm": 0.26470532087807114,
+      "learning_rate": 2.9831397944888833e-05,
+      "loss": 0.5586,
+      "step": 259
+    },
+    {
+      "epoch": 0.46973803071364045,
+      "grad_norm": 0.244315447384743,
+      "learning_rate": 2.9684532864643122e-05,
+      "loss": 0.5719,
+      "step": 260
+    },
+    {
+      "epoch": 0.4715447154471545,
+      "grad_norm": 0.2579290960374767,
+      "learning_rate": 2.953750004073041e-05,
+      "loss": 0.5625,
+      "step": 261
+    },
+    {
+      "epoch": 0.47335140018066846,
+      "grad_norm": 0.2502355307959324,
+      "learning_rate": 2.9390304738099384e-05,
+      "loss": 0.5508,
+      "step": 262
+    },
+    {
+      "epoch": 0.4751580849141825,
+      "grad_norm": 0.22745648145569985,
+      "learning_rate": 2.9242952227516722e-05,
+      "loss": 0.5558,
+      "step": 263
+    },
+    {
+      "epoch": 0.47696476964769646,
+      "grad_norm": 0.2880551827736454,
+      "learning_rate": 2.9095447785378443e-05,
+      "loss": 0.5615,
+      "step": 264
+    },
+    {
+      "epoch": 0.4787714543812105,
+      "grad_norm": 0.21091165839321852,
+      "learning_rate": 2.89477966935209e-05,
+      "loss": 0.543,
+      "step": 265
+    },
+    {
+      "epoch": 0.48057813911472447,
+      "grad_norm": 0.3146456092706863,
+      "learning_rate": 2.8800004239031684e-05,
+      "loss": 0.5676,
+      "step": 266
+    },
+    {
+      "epoch": 0.4823848238482385,
+      "grad_norm": 0.2187643898145053,
+      "learning_rate": 2.8652075714060295e-05,
+      "loss": 0.552,
+      "step": 267
+    },
+    {
+      "epoch": 0.48419150858175247,
+      "grad_norm": 0.25789704379288414,
+      "learning_rate": 2.850401641562865e-05,
+      "loss": 0.5511,
+      "step": 268
+    },
+    {
+      "epoch": 0.4859981933152665,
+      "grad_norm": 0.23735131054824035,
+      "learning_rate": 2.8355831645441388e-05,
+      "loss": 0.5593,
+      "step": 269
+    },
+    {
+      "epoch": 0.4878048780487805,
+      "grad_norm": 0.22178545968987512,
+      "learning_rate": 2.8207526709696057e-05,
+      "loss": 0.54,
+      "step": 270
+    },
+    {
+      "epoch": 0.4896115627822945,
+      "grad_norm": 0.2444983259488006,
+      "learning_rate": 2.8059106918893068e-05,
+      "loss": 0.5618,
+      "step": 271
+    },
+    {
+      "epoch": 0.4914182475158085,
+      "grad_norm": 0.23074078288504987,
+      "learning_rate": 2.791057758764557e-05,
+      "loss": 0.5561,
+      "step": 272
+    },
+    {
+      "epoch": 0.4932249322493225,
+      "grad_norm": 0.2122648915326668,
+      "learning_rate": 2.7761944034489152e-05,
+      "loss": 0.564,
+      "step": 273
+    },
+    {
+      "epoch": 0.4950316169828365,
+      "grad_norm": 0.21492258086562654,
+      "learning_rate": 2.761321158169134e-05,
+      "loss": 0.5517,
+      "step": 274
+    },
+    {
+      "epoch": 0.4968383017163505,
+      "grad_norm": 0.27472351877781365,
+      "learning_rate": 2.746438555506109e-05,
+      "loss": 0.5462,
+      "step": 275
+    },
+    {
+      "epoch": 0.4986449864498645,
+      "grad_norm": 0.21428421421146693,
+      "learning_rate": 2.7315471283758037e-05,
+      "loss": 0.5378,
+      "step": 276
+    },
+    {
+      "epoch": 0.5004516711833785,
+      "grad_norm": 0.22502606248228424,
+      "learning_rate": 2.7166474100101673e-05,
+      "loss": 0.5497,
+      "step": 277
+    },
+    {
+      "epoch": 0.5022583559168925,
+      "grad_norm": 0.22837843189987042,
+      "learning_rate": 2.7017399339380434e-05,
+      "loss": 0.5371,
+      "step": 278
+    },
+    {
+      "epoch": 0.5040650406504065,
+      "grad_norm": 0.24309416552693688,
+      "learning_rate": 2.686825233966061e-05,
+      "loss": 0.5521,
+      "step": 279
+    },
+    {
+      "epoch": 0.5058717253839206,
+      "grad_norm": 0.21900766697042293,
+      "learning_rate": 2.6719038441595233e-05,
+      "loss": 0.5481,
+      "step": 280
+    },
+    {
+      "epoch": 0.5076784101174345,
+      "grad_norm": 0.22388193391843347,
+      "learning_rate": 2.656976298823284e-05,
+      "loss": 0.5491,
+      "step": 281
+    },
+    {
+      "epoch": 0.5094850948509485,
+      "grad_norm": 0.22518528793598785,
+      "learning_rate": 2.6420431324826117e-05,
+      "loss": 0.5437,
+      "step": 282
+    },
+    {
+      "epoch": 0.5112917795844625,
+      "grad_norm": 0.22899910230756002,
+      "learning_rate": 2.6271048798640547e-05,
+      "loss": 0.5585,
+      "step": 283
+    },
+    {
+      "epoch": 0.5130984643179766,
+      "grad_norm": 0.23580513860894517,
+      "learning_rate": 2.6121620758762877e-05,
+      "loss": 0.5466,
+      "step": 284
+    },
+    {
+      "epoch": 0.5149051490514905,
+      "grad_norm": 0.2424173649022125,
+      "learning_rate": 2.5972152555909625e-05,
+      "loss": 0.542,
+      "step": 285
+    },
+    {
+      "epoch": 0.5167118337850045,
+      "grad_norm": 0.21694029342367355,
+      "learning_rate": 2.5822649542235466e-05,
+      "loss": 0.5386,
+      "step": 286
+    },
+    {
+      "epoch": 0.5185185185185185,
+      "grad_norm": 0.25827875560858626,
+      "learning_rate": 2.5673117071141572e-05,
+      "loss": 0.5601,
+      "step": 287
+    },
+    {
+      "epoch": 0.5203252032520326,
+      "grad_norm": 0.23262800077998166,
+      "learning_rate": 2.5523560497083926e-05,
+      "loss": 0.5575,
+      "step": 288
+    },
+    {
+      "epoch": 0.5221318879855466,
+      "grad_norm": 0.2761966922116131,
+      "learning_rate": 2.5373985175381594e-05,
+      "loss": 0.5418,
+      "step": 289
+    },
+    {
+      "epoch": 0.5239385727190605,
+      "grad_norm": 0.26156166020108806,
+      "learning_rate": 2.5224396462024947e-05,
+      "loss": 0.5539,
+      "step": 290
+    },
+    {
+      "epoch": 0.5257452574525745,
+      "grad_norm": 0.21779581948691018,
+      "learning_rate": 2.507479971348391e-05,
+      "loss": 0.5581,
+      "step": 291
+    },
+    {
+      "epoch": 0.5275519421860885,
+      "grad_norm": 0.2377723759423303,
+      "learning_rate": 2.4925200286516097e-05,
+      "loss": 0.5621,
+      "step": 292
+    },
+    {
+      "epoch": 0.5293586269196026,
+      "grad_norm": 0.25645028358412414,
+      "learning_rate": 2.4775603537975052e-05,
+      "loss": 0.5545,
+      "step": 293
+    },
+    {
+      "epoch": 0.5311653116531165,
+      "grad_norm": 0.23027125665344333,
+      "learning_rate": 2.4626014824618415e-05,
+      "loss": 0.5466,
+      "step": 294
+    },
+    {
+      "epoch": 0.5329719963866305,
+      "grad_norm": 0.21506264402370667,
+      "learning_rate": 2.447643950291608e-05,
+      "loss": 0.5444,
+      "step": 295
+    },
+    {
+      "epoch": 0.5347786811201445,
+      "grad_norm": 0.23689584066775435,
+      "learning_rate": 2.4326882928858434e-05,
+      "loss": 0.5486,
+      "step": 296
+    },
+    {
+      "epoch": 0.5365853658536586,
+      "grad_norm": 0.24755488327450098,
+      "learning_rate": 2.417735045776453e-05,
+      "loss": 0.553,
+      "step": 297
+    },
+    {
+      "epoch": 0.5383920505871725,
+      "grad_norm": 0.23203698871403608,
+      "learning_rate": 2.402784744409038e-05,
+      "loss": 0.5484,
+      "step": 298
+    },
+    {
+      "epoch": 0.5401987353206865,
+      "grad_norm": 0.24571656282529886,
+      "learning_rate": 2.3878379241237136e-05,
+      "loss": 0.5632,
+      "step": 299
+    },
+    {
+      "epoch": 0.5420054200542005,
+      "grad_norm": 0.20706812965056257,
+      "learning_rate": 2.372895120135946e-05,
+      "loss": 0.5513,
+      "step": 300
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 553,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 347755278106624.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3232dab23c17d3f385c8ad2855e4120afbef5562c13a830c5fcc7469b187f32b
+size 8081