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:fe6a3162f8731b23c1f14d15f750c3093313da56fbd09e913a1fae98ab402b40
+size 4997184968

model-00002-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf43b21cdb15eab0552fecec53dcac116c91820fd9cdbaa4a6551204a0252a43
+size 4997741608

model-00003-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e0b4f230eb5921d51fc69b628883907edb01b34c306b8b6cd6f8f70dbb97854
+size 4997742208

model-00004-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c24277ea669d3f01fea07ca2dd482a5af3743ffb8c9ff3606253d0ce0930596
+size 4997743184

model-00005-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5d3ec712f6f68a7b63a479554df85de2a6a9cf08af37436117bbf19833b7289
+size 4997743184

model-00006-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4008460fe013781e622c251eced36c205ccf309f29f8f7c1a299ce2da385184e
+size 4997743184

model-00007-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f4bd90921e290c69cb47d7e0537d3b0b222f3190e4458f02432cb4082e597ff
+size 4997743184

model-00008-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b43cfe4f5fa0a18f172b740dd2ce22069cdb7545806846f60991aeae7f5f4a4
+size 4997743184

model-00009-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6f6b5d95ce88ab4e08f57a143488a4e6ab7ef5ae228e0c1a5ef32522c8485e1
+size 4997743184

model-00010-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7991dd51bbf60c0ff951373d8f7f829c1bfa5c65605011f4dfdf8c670c0f2d54
+size 4997743184

model-00011-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b127dcbd04f5db71d46a0564c81ac1c8ea8d51bedbc8022d9f9ccbd3b5fea8da
+size 4997743184

model-00012-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2f6f6f14ae7c411d6ac1df40203a247c780131d318c87c340077ddd62c793d3
+size 4997743184

model-00013-of-00013.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:892366e0ec9182e13d66a8a595c1d7b04f4c01871ac45cf2d10606f667b3305b
+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 +1,2232 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.6666666666666666,
+  "eval_steps": 500,
+  "global_step": 314,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0021231422505307855,
+      "grad_norm": 6.233692311689662,
+      "learning_rate": 0.0,
+      "loss": 1.3677,
+      "step": 1
+    },
+    {
+      "epoch": 0.004246284501061571,
+      "grad_norm": 6.0364996372518975,
+      "learning_rate": 1.2499999999999999e-06,
+      "loss": 1.4092,
+      "step": 2
+    },
+    {
+      "epoch": 0.006369426751592357,
+      "grad_norm": 5.824376154111525,
+      "learning_rate": 2.4999999999999998e-06,
+      "loss": 1.3994,
+      "step": 3
+    },
+    {
+      "epoch": 0.008492569002123142,
+      "grad_norm": 5.902606326788258,
+      "learning_rate": 3.75e-06,
+      "loss": 1.3973,
+      "step": 4
+    },
+    {
+      "epoch": 0.010615711252653927,
+      "grad_norm": 5.275994860996141,
+      "learning_rate": 4.9999999999999996e-06,
+      "loss": 1.3653,
+      "step": 5
+    },
+    {
+      "epoch": 0.012738853503184714,
+      "grad_norm": 4.1623214153554295,
+      "learning_rate": 6.25e-06,
+      "loss": 1.3567,
+      "step": 6
+    },
+    {
+      "epoch": 0.014861995753715499,
+      "grad_norm": 3.3671066127082088,
+      "learning_rate": 7.5e-06,
+      "loss": 1.2129,
+      "step": 7
+    },
+    {
+      "epoch": 0.016985138004246284,
+      "grad_norm": 2.97534257184618,
+      "learning_rate": 8.750000000000001e-06,
+      "loss": 1.2794,
+      "step": 8
+    },
+    {
+      "epoch": 0.01910828025477707,
+      "grad_norm": 2.9074617656297654,
+      "learning_rate": 9.999999999999999e-06,
+      "loss": 1.2212,
+      "step": 9
+    },
+    {
+      "epoch": 0.021231422505307854,
+      "grad_norm": 3.829816734428447,
+      "learning_rate": 1.125e-05,
+      "loss": 1.2232,
+      "step": 10
+    },
+    {
+      "epoch": 0.02335456475583864,
+      "grad_norm": 2.710794346051666,
+      "learning_rate": 1.25e-05,
+      "loss": 1.1805,
+      "step": 11
+    },
+    {
+      "epoch": 0.025477707006369428,
+      "grad_norm": 2.490537571865269,
+      "learning_rate": 1.375e-05,
+      "loss": 1.1395,
+      "step": 12
+    },
+    {
+      "epoch": 0.027600849256900213,
+      "grad_norm": 2.958343893823127,
+      "learning_rate": 1.5e-05,
+      "loss": 1.1996,
+      "step": 13
+    },
+    {
+      "epoch": 0.029723991507430998,
+      "grad_norm": 2.10711525179354,
+      "learning_rate": 1.625e-05,
+      "loss": 1.1154,
+      "step": 14
+    },
+    {
+      "epoch": 0.03184713375796178,
+      "grad_norm": 1.804171701556794,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 1.1335,
+      "step": 15
+    },
+    {
+      "epoch": 0.03397027600849257,
+      "grad_norm": 1.3531726745218633,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 1.0982,
+      "step": 16
+    },
+    {
+      "epoch": 0.036093418259023353,
+      "grad_norm": 1.2822932286416207,
+      "learning_rate": 1.9999999999999998e-05,
+      "loss": 1.0626,
+      "step": 17
+    },
+    {
+      "epoch": 0.03821656050955414,
+      "grad_norm": 1.2087880979605115,
+      "learning_rate": 2.125e-05,
+      "loss": 1.0372,
+      "step": 18
+    },
+    {
+      "epoch": 0.040339702760084924,
+      "grad_norm": 1.2492451284743782,
+      "learning_rate": 2.25e-05,
+      "loss": 1.0343,
+      "step": 19
+    },
+    {
+      "epoch": 0.04246284501061571,
+      "grad_norm": 1.10733509517002,
+      "learning_rate": 2.3749999999999998e-05,
+      "loss": 0.9442,
+      "step": 20
+    },
+    {
+      "epoch": 0.044585987261146494,
+      "grad_norm": 1.0490035536866515,
+      "learning_rate": 2.5e-05,
+      "loss": 1.0626,
+      "step": 21
+    },
+    {
+      "epoch": 0.04670912951167728,
+      "grad_norm": 0.9466743921266996,
+      "learning_rate": 2.625e-05,
+      "loss": 1.0386,
+      "step": 22
+    },
+    {
+      "epoch": 0.04883227176220807,
+      "grad_norm": 0.9081034267931447,
+      "learning_rate": 2.75e-05,
+      "loss": 0.9487,
+      "step": 23
+    },
+    {
+      "epoch": 0.050955414012738856,
+      "grad_norm": 1.0372847960112026,
+      "learning_rate": 2.875e-05,
+      "loss": 0.9728,
+      "step": 24
+    },
+    {
+      "epoch": 0.05307855626326964,
+      "grad_norm": 1.198809452316821,
+      "learning_rate": 3e-05,
+      "loss": 1.0076,
+      "step": 25
+    },
+    {
+      "epoch": 0.055201698513800426,
+      "grad_norm": 1.1303411718578629,
+      "learning_rate": 2.9999629537566803e-05,
+      "loss": 1.0052,
+      "step": 26
+    },
+    {
+      "epoch": 0.05732484076433121,
+      "grad_norm": 0.9253246602948907,
+      "learning_rate": 2.9998518168566207e-05,
+      "loss": 1.0535,
+      "step": 27
+    },
+    {
+      "epoch": 0.059447983014861996,
+      "grad_norm": 0.8916584736775167,
+      "learning_rate": 2.999666594789427e-05,
+      "loss": 1.0575,
+      "step": 28
+    },
+    {
+      "epoch": 0.06157112526539278,
+      "grad_norm": 1.017365856913298,
+      "learning_rate": 2.999407296704142e-05,
+      "loss": 1.0276,
+      "step": 29
+    },
+    {
+      "epoch": 0.06369426751592357,
+      "grad_norm": 0.9179032077950415,
+      "learning_rate": 2.9990739354087918e-05,
+      "loss": 1.0334,
+      "step": 30
+    },
+    {
+      "epoch": 0.06581740976645435,
+      "grad_norm": 0.9094459507546309,
+      "learning_rate": 2.9986665273697548e-05,
+      "loss": 1.0191,
+      "step": 31
+    },
+    {
+      "epoch": 0.06794055201698514,
+      "grad_norm": 0.8328888734598474,
+      "learning_rate": 2.9981850927109472e-05,
+      "loss": 1.0266,
+      "step": 32
+    },
+    {
+      "epoch": 0.07006369426751592,
+      "grad_norm": 0.9265287035853829,
+      "learning_rate": 2.9976296552128305e-05,
+      "loss": 0.9839,
+      "step": 33
+    },
+    {
+      "epoch": 0.07218683651804671,
+      "grad_norm": 0.8386673953378151,
+      "learning_rate": 2.9970002423112342e-05,
+      "loss": 0.9856,
+      "step": 34
+    },
+    {
+      "epoch": 0.07430997876857749,
+      "grad_norm": 0.8773223216488372,
+      "learning_rate": 2.996296885096003e-05,
+      "loss": 0.9913,
+      "step": 35
+    },
+    {
+      "epoch": 0.07643312101910828,
+      "grad_norm": 1.9920269675167115,
+      "learning_rate": 2.9955196183094604e-05,
+      "loss": 1.0276,
+      "step": 36
+    },
+    {
+      "epoch": 0.07855626326963906,
+      "grad_norm": 0.8541287848004893,
+      "learning_rate": 2.9946684803446928e-05,
+      "loss": 0.9692,
+      "step": 37
+    },
+    {
+      "epoch": 0.08067940552016985,
+      "grad_norm": 0.8475286973659103,
+      "learning_rate": 2.9937435132436517e-05,
+      "loss": 0.9648,
+      "step": 38
+    },
+    {
+      "epoch": 0.08280254777070063,
+      "grad_norm": 0.891470065700579,
+      "learning_rate": 2.9927447626950795e-05,
+      "loss": 0.9469,
+      "step": 39
+    },
+    {
+      "epoch": 0.08492569002123142,
+      "grad_norm": 0.918643932520895,
+      "learning_rate": 2.9916722780322504e-05,
+      "loss": 1.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.0870488322717622,
+      "grad_norm": 0.89952358144301,
+      "learning_rate": 2.9905261122305344e-05,
+      "loss": 0.9591,
+      "step": 41
+    },
+    {
+      "epoch": 0.08917197452229299,
+      "grad_norm": 0.9534501125480317,
+      "learning_rate": 2.9893063219047815e-05,
+      "loss": 0.9638,
+      "step": 42
+    },
+    {
+      "epoch": 0.09129511677282377,
+      "grad_norm": 0.9410259599712866,
+      "learning_rate": 2.988012967306524e-05,
+      "loss": 0.9631,
+      "step": 43
+    },
+    {
+      "epoch": 0.09341825902335456,
+      "grad_norm": 0.7903964300371441,
+      "learning_rate": 2.9866461123210005e-05,
+      "loss": 1.0352,
+      "step": 44
+    },
+    {
+      "epoch": 0.09554140127388536,
+      "grad_norm": 0.8186703306518206,
+      "learning_rate": 2.985205824464001e-05,
+      "loss": 1.035,
+      "step": 45
+    },
+    {
+      "epoch": 0.09766454352441614,
+      "grad_norm": 1.1828963478090526,
+      "learning_rate": 2.983692174878531e-05,
+      "loss": 1.0224,
+      "step": 46
+    },
+    {
+      "epoch": 0.09978768577494693,
+      "grad_norm": 1.0653437359039768,
+      "learning_rate": 2.9821052383312987e-05,
+      "loss": 0.9845,
+      "step": 47
+    },
+    {
+      "epoch": 0.10191082802547771,
+      "grad_norm": 0.9139888799070773,
+      "learning_rate": 2.980445093209021e-05,
+      "loss": 0.9882,
+      "step": 48
+    },
+    {
+      "epoch": 0.1040339702760085,
+      "grad_norm": 1.0047461469584085,
+      "learning_rate": 2.9787118215145502e-05,
+      "loss": 0.9707,
+      "step": 49
+    },
+    {
+      "epoch": 0.10615711252653928,
+      "grad_norm": 0.9419818838483188,
+      "learning_rate": 2.976905508862828e-05,
+      "loss": 1.0093,
+      "step": 50
+    },
+    {
+      "epoch": 0.10828025477707007,
+      "grad_norm": 0.8165141706193597,
+      "learning_rate": 2.9750262444766502e-05,
+      "loss": 0.9982,
+      "step": 51
+    },
+    {
+      "epoch": 0.11040339702760085,
+      "grad_norm": 0.9276179840515267,
+      "learning_rate": 2.9730741211822654e-05,
+      "loss": 1.0062,
+      "step": 52
+    },
+    {
+      "epoch": 0.11252653927813164,
+      "grad_norm": 0.7769869987656481,
+      "learning_rate": 2.9710492354047857e-05,
+      "loss": 0.9344,
+      "step": 53
+    },
+    {
+      "epoch": 0.11464968152866242,
+      "grad_norm": 0.8691988768513328,
+      "learning_rate": 2.968951687163426e-05,
+      "loss": 0.9953,
+      "step": 54
+    },
+    {
+      "epoch": 0.11677282377919321,
+      "grad_norm": 0.8209469205962703,
+      "learning_rate": 2.9667815800665637e-05,
+      "loss": 1.0306,
+      "step": 55
+    },
+    {
+      "epoch": 0.11889596602972399,
+      "grad_norm": 0.7477038589692927,
+      "learning_rate": 2.9645390213066193e-05,
+      "loss": 1.0604,
+      "step": 56
+    },
+    {
+      "epoch": 0.12101910828025478,
+      "grad_norm": 0.8205605966500025,
+      "learning_rate": 2.9622241216547623e-05,
+      "loss": 0.9994,
+      "step": 57
+    },
+    {
+      "epoch": 0.12314225053078556,
+      "grad_norm": 0.8004707255796194,
+      "learning_rate": 2.9598369954554405e-05,
+      "loss": 0.9995,
+      "step": 58
+    },
+    {
+      "epoch": 0.12526539278131635,
+      "grad_norm": 0.7081299100114782,
+      "learning_rate": 2.957377760620732e-05,
+      "loss": 0.9943,
+      "step": 59
+    },
+    {
+      "epoch": 0.12738853503184713,
+      "grad_norm": 0.7217649291059394,
+      "learning_rate": 2.9548465386245185e-05,
+      "loss": 0.987,
+      "step": 60
+    },
+    {
+      "epoch": 0.12951167728237792,
+      "grad_norm": 0.8205996950485613,
+      "learning_rate": 2.952243454496488e-05,
+      "loss": 0.9136,
+      "step": 61
+    },
+    {
+      "epoch": 0.1316348195329087,
+      "grad_norm": 0.7201961172816332,
+      "learning_rate": 2.9495686368159592e-05,
+      "loss": 0.976,
+      "step": 62
+    },
+    {
+      "epoch": 0.1337579617834395,
+      "grad_norm": 0.7696946636431424,
+      "learning_rate": 2.946822217705526e-05,
+      "loss": 0.9993,
+      "step": 63
+    },
+    {
+      "epoch": 0.13588110403397027,
+      "grad_norm": 0.7848756346253257,
+      "learning_rate": 2.9440043328245366e-05,
+      "loss": 1.0206,
+      "step": 64
+    },
+    {
+      "epoch": 0.13800424628450106,
+      "grad_norm": 0.740850768759331,
+      "learning_rate": 2.9411151213623894e-05,
+      "loss": 1.0311,
+      "step": 65
+    },
+    {
+      "epoch": 0.14012738853503184,
+      "grad_norm": 0.7028782098692511,
+      "learning_rate": 2.938154726031659e-05,
+      "loss": 0.9681,
+      "step": 66
+    },
+    {
+      "epoch": 0.14225053078556263,
+      "grad_norm": 0.7087905746550794,
+      "learning_rate": 2.9351232930610473e-05,
+      "loss": 0.975,
+      "step": 67
+    },
+    {
+      "epoch": 0.14437367303609341,
+      "grad_norm": 0.7512258835012833,
+      "learning_rate": 2.932020972188157e-05,
+      "loss": 0.9703,
+      "step": 68
+    },
+    {
+      "epoch": 0.1464968152866242,
+      "grad_norm": 0.6741016163735575,
+      "learning_rate": 2.9288479166521014e-05,
+      "loss": 0.9495,
+      "step": 69
+    },
+    {
+      "epoch": 0.14861995753715498,
+      "grad_norm": 0.740748414113653,
+      "learning_rate": 2.9256042831859295e-05,
+      "loss": 1.0603,
+      "step": 70
+    },
+    {
+      "epoch": 0.15074309978768577,
+      "grad_norm": 0.7918692932011524,
+      "learning_rate": 2.9222902320088882e-05,
+      "loss": 1.0028,
+      "step": 71
+    },
+    {
+      "epoch": 0.15286624203821655,
+      "grad_norm": 0.759432829778867,
+      "learning_rate": 2.9189059268185057e-05,
+      "loss": 1.0205,
+      "step": 72
+    },
+    {
+      "epoch": 0.15498938428874734,
+      "grad_norm": 0.7025218211620403,
+      "learning_rate": 2.9154515347825065e-05,
+      "loss": 0.9696,
+      "step": 73
+    },
+    {
+      "epoch": 0.15711252653927812,
+      "grad_norm": 0.8396244762201516,
+      "learning_rate": 2.9119272265305546e-05,
+      "loss": 0.9472,
+      "step": 74
+    },
+    {
+      "epoch": 0.1592356687898089,
+      "grad_norm": 0.7538155851518231,
+      "learning_rate": 2.9083331761458247e-05,
+      "loss": 0.992,
+      "step": 75
+    },
+    {
+      "epoch": 0.1613588110403397,
+      "grad_norm": 0.8141973691781073,
+      "learning_rate": 2.904669561156404e-05,
+      "loss": 0.9097,
+      "step": 76
+    },
+    {
+      "epoch": 0.16348195329087048,
+      "grad_norm": 0.6587425676342493,
+      "learning_rate": 2.9009365625265215e-05,
+      "loss": 0.9604,
+      "step": 77
+    },
+    {
+      "epoch": 0.16560509554140126,
+      "grad_norm": 0.7407423574470945,
+      "learning_rate": 2.8971343646476114e-05,
+      "loss": 0.9779,
+      "step": 78
+    },
+    {
+      "epoch": 0.16772823779193205,
+      "grad_norm": 0.8004547291076868,
+      "learning_rate": 2.893263155329204e-05,
+      "loss": 0.9766,
+      "step": 79
+    },
+    {
+      "epoch": 0.16985138004246284,
+      "grad_norm": 0.7457062213618112,
+      "learning_rate": 2.8893231257896502e-05,
+      "loss": 1.0744,
+      "step": 80
+    },
+    {
+      "epoch": 0.17197452229299362,
+      "grad_norm": 0.7225596471074269,
+      "learning_rate": 2.8853144706466725e-05,
+      "loss": 1.0006,
+      "step": 81
+    },
+    {
+      "epoch": 0.1740976645435244,
+      "grad_norm": 0.9234525636361333,
+      "learning_rate": 2.881237387907757e-05,
+      "loss": 1.0595,
+      "step": 82
+    },
+    {
+      "epoch": 0.1762208067940552,
+      "grad_norm": 0.7277875089674879,
+      "learning_rate": 2.8770920789603687e-05,
+      "loss": 0.9764,
+      "step": 83
+    },
+    {
+      "epoch": 0.17834394904458598,
+      "grad_norm": 0.7251182820248202,
+      "learning_rate": 2.8728787485620068e-05,
+      "loss": 1.0087,
+      "step": 84
+    },
+    {
+      "epoch": 0.18046709129511676,
+      "grad_norm": 0.7259248063014598,
+      "learning_rate": 2.868597604830088e-05,
+      "loss": 0.9895,
+      "step": 85
+    },
+    {
+      "epoch": 0.18259023354564755,
+      "grad_norm": 0.7236366037154788,
+      "learning_rate": 2.864248859231669e-05,
+      "loss": 1.0186,
+      "step": 86
+    },
+    {
+      "epoch": 0.18471337579617833,
+      "grad_norm": 0.7104922685407246,
+      "learning_rate": 2.859832726573e-05,
+      "loss": 1.0179,
+      "step": 87
+    },
+    {
+      "epoch": 0.18683651804670912,
+      "grad_norm": 0.6494155954698466,
+      "learning_rate": 2.855349424988915e-05,
+      "loss": 0.9532,
+      "step": 88
+    },
+    {
+      "epoch": 0.18895966029723993,
+      "grad_norm": 0.7642728344785427,
+      "learning_rate": 2.8507991759320545e-05,
+      "loss": 0.931,
+      "step": 89
+    },
+    {
+      "epoch": 0.1910828025477707,
+      "grad_norm": 0.7579534251967283,
+      "learning_rate": 2.8461822041619312e-05,
+      "loss": 1.0023,
+      "step": 90
+    },
+    {
+      "epoch": 0.1932059447983015,
+      "grad_norm": 0.700517970884762,
+      "learning_rate": 2.841498737733824e-05,
+      "loss": 0.9251,
+      "step": 91
+    },
+    {
+      "epoch": 0.19532908704883228,
+      "grad_norm": 0.7550458532390354,
+      "learning_rate": 2.8367490079875154e-05,
+      "loss": 0.9632,
+      "step": 92
+    },
+    {
+      "epoch": 0.19745222929936307,
+      "grad_norm": 0.7561266680050324,
+      "learning_rate": 2.831933249535865e-05,
+      "loss": 0.975,
+      "step": 93
+    },
+    {
+      "epoch": 0.19957537154989385,
+      "grad_norm": 0.7554339394976815,
+      "learning_rate": 2.827051700253217e-05,
+      "loss": 0.9574,
+      "step": 94
+    },
+    {
+      "epoch": 0.20169851380042464,
+      "grad_norm": 0.7519417055562898,
+      "learning_rate": 2.8221046012636558e-05,
+      "loss": 1.0089,
+      "step": 95
+    },
+    {
+      "epoch": 0.20382165605095542,
+      "grad_norm": 0.695300221944843,
+      "learning_rate": 2.817092196929091e-05,
+      "loss": 0.9586,
+      "step": 96
+    },
+    {
+      "epoch": 0.2059447983014862,
+      "grad_norm": 0.7004398610006882,
+      "learning_rate": 2.8120147348371912e-05,
+      "loss": 0.9575,
+      "step": 97
+    },
+    {
+      "epoch": 0.208067940552017,
+      "grad_norm": 0.7161258126899045,
+      "learning_rate": 2.8068724657891507e-05,
+      "loss": 0.9429,
+      "step": 98
+    },
+    {
+      "epoch": 0.21019108280254778,
+      "grad_norm": 0.7498697224476674,
+      "learning_rate": 2.801665643787303e-05,
+      "loss": 1.0132,
+      "step": 99
+    },
+    {
+      "epoch": 0.21231422505307856,
+      "grad_norm": 0.6820422896195972,
+      "learning_rate": 2.7963945260225748e-05,
+      "loss": 1.0142,
+      "step": 100
+    },
+    {
+      "epoch": 0.21443736730360935,
+      "grad_norm": 0.6998151084514577,
+      "learning_rate": 2.7910593728617813e-05,
+      "loss": 1.0474,
+      "step": 101
+    },
+    {
+      "epoch": 0.21656050955414013,
+      "grad_norm": 0.6659748871284866,
+      "learning_rate": 2.7856604478347655e-05,
+      "loss": 0.964,
+      "step": 102
+    },
+    {
+      "epoch": 0.21868365180467092,
+      "grad_norm": 0.6639674612741819,
+      "learning_rate": 2.7801980176213798e-05,
+      "loss": 0.9141,
+      "step": 103
+    },
+    {
+      "epoch": 0.2208067940552017,
+      "grad_norm": 0.7962940595848361,
+      "learning_rate": 2.7746723520383174e-05,
+      "loss": 0.98,
+      "step": 104
+    },
+    {
+      "epoch": 0.2229299363057325,
+      "grad_norm": 0.6748458306743221,
+      "learning_rate": 2.76908372402578e-05,
+      "loss": 0.9387,
+      "step": 105
+    },
+    {
+      "epoch": 0.22505307855626328,
+      "grad_norm": 0.6591397874457393,
+      "learning_rate": 2.763432409633998e-05,
+      "loss": 0.9401,
+      "step": 106
+    },
+    {
+      "epoch": 0.22717622080679406,
+      "grad_norm": 0.6815163810802223,
+      "learning_rate": 2.7577186880095966e-05,
+      "loss": 0.9852,
+      "step": 107
+    },
+    {
+      "epoch": 0.22929936305732485,
+      "grad_norm": 0.6786470466077741,
+      "learning_rate": 2.7519428413818034e-05,
+      "loss": 0.9549,
+      "step": 108
+    },
+    {
+      "epoch": 0.23142250530785563,
+      "grad_norm": 0.722955020585509,
+      "learning_rate": 2.7461051550485116e-05,
+      "loss": 1.0021,
+      "step": 109
+    },
+    {
+      "epoch": 0.23354564755838642,
+      "grad_norm": 0.7518724654809783,
+      "learning_rate": 2.740205917362186e-05,
+      "loss": 1.0168,
+      "step": 110
+    },
+    {
+      "epoch": 0.2356687898089172,
+      "grad_norm": 0.5989515308193117,
+      "learning_rate": 2.7342454197156194e-05,
+      "loss": 0.9608,
+      "step": 111
+    },
+    {
+      "epoch": 0.23779193205944799,
+      "grad_norm": 0.63739062901645,
+      "learning_rate": 2.728223956527539e-05,
+      "loss": 0.9779,
+      "step": 112
+    },
+    {
+      "epoch": 0.23991507430997877,
+      "grad_norm": 0.7107030178119403,
+      "learning_rate": 2.722141825228066e-05,
+      "loss": 0.9507,
+      "step": 113
+    },
+    {
+      "epoch": 0.24203821656050956,
+      "grad_norm": 0.6662094575605222,
+      "learning_rate": 2.7159993262440228e-05,
+      "loss": 0.9566,
+      "step": 114
+    },
+    {
+      "epoch": 0.24416135881104034,
+      "grad_norm": 0.684949594473451,
+      "learning_rate": 2.7097967629840906e-05,
+      "loss": 1.0072,
+      "step": 115
+    },
+    {
+      "epoch": 0.24628450106157113,
+      "grad_norm": 0.717234854863913,
+      "learning_rate": 2.703534441823827e-05,
+      "loss": 0.9774,
+      "step": 116
+    },
+    {
+      "epoch": 0.2484076433121019,
+      "grad_norm": 0.6600997556047009,
+      "learning_rate": 2.6972126720905293e-05,
+      "loss": 0.9804,
+      "step": 117
+    },
+    {
+      "epoch": 0.2505307855626327,
+      "grad_norm": 0.6717502933643555,
+      "learning_rate": 2.6908317660479583e-05,
+      "loss": 0.9805,
+      "step": 118
+    },
+    {
+      "epoch": 0.2526539278131635,
+      "grad_norm": 0.6668903744774365,
+      "learning_rate": 2.68439203888091e-05,
+      "loss": 1.012,
+      "step": 119
+    },
+    {
+      "epoch": 0.25477707006369427,
+      "grad_norm": 0.7423058243771732,
+      "learning_rate": 2.6778938086796512e-05,
+      "loss": 1.0772,
+      "step": 120
+    },
+    {
+      "epoch": 0.25690021231422505,
+      "grad_norm": 0.7186526888110079,
+      "learning_rate": 2.6713373964242043e-05,
+      "loss": 0.9614,
+      "step": 121
+    },
+    {
+      "epoch": 0.25902335456475584,
+      "grad_norm": 0.7020168430150423,
+      "learning_rate": 2.6647231259684946e-05,
+      "loss": 0.9639,
+      "step": 122
+    },
+    {
+      "epoch": 0.2611464968152866,
+      "grad_norm": 0.6506699216795314,
+      "learning_rate": 2.6580513240243524e-05,
+      "loss": 1.0157,
+      "step": 123
+    },
+    {
+      "epoch": 0.2632696390658174,
+      "grad_norm": 0.6979496800781052,
+      "learning_rate": 2.651322320145375e-05,
+      "loss": 0.9827,
+      "step": 124
+    },
+    {
+      "epoch": 0.2653927813163482,
+      "grad_norm": 0.7041593201924282,
+      "learning_rate": 2.6445364467106494e-05,
+      "loss": 0.9416,
+      "step": 125
+    },
+    {
+      "epoch": 0.267515923566879,
+      "grad_norm": 0.7474469108757689,
+      "learning_rate": 2.637694038908333e-05,
+      "loss": 1.0345,
+      "step": 126
+    },
+    {
+      "epoch": 0.26963906581740976,
+      "grad_norm": 0.6614249491962885,
+      "learning_rate": 2.630795434719099e-05,
+      "loss": 0.9453,
+      "step": 127
+    },
+    {
+      "epoch": 0.27176220806794055,
+      "grad_norm": 0.6657326589669448,
+      "learning_rate": 2.623840974899439e-05,
+      "loss": 0.889,
+      "step": 128
+    },
+    {
+      "epoch": 0.27388535031847133,
+      "grad_norm": 0.8022831021638541,
+      "learning_rate": 2.616831002964834e-05,
+      "loss": 0.9739,
+      "step": 129
+    },
+    {
+      "epoch": 0.2760084925690021,
+      "grad_norm": 0.7179257591811719,
+      "learning_rate": 2.609765865172786e-05,
+      "loss": 0.9843,
+      "step": 130
+    },
+    {
+      "epoch": 0.2781316348195329,
+      "grad_norm": 0.605816781626969,
+      "learning_rate": 2.6026459105057127e-05,
+      "loss": 0.96,
+      "step": 131
+    },
+    {
+      "epoch": 0.2802547770700637,
+      "grad_norm": 0.6689266661628013,
+      "learning_rate": 2.5954714906537116e-05,
+      "loss": 0.9829,
+      "step": 132
+    },
+    {
+      "epoch": 0.2823779193205945,
+      "grad_norm": 0.6414905131754337,
+      "learning_rate": 2.5882429599971872e-05,
+      "loss": 0.9315,
+      "step": 133
+    },
+    {
+      "epoch": 0.28450106157112526,
+      "grad_norm": 0.6383947258042313,
+      "learning_rate": 2.580960675589347e-05,
+      "loss": 1.0047,
+      "step": 134
+    },
+    {
+      "epoch": 0.28662420382165604,
+      "grad_norm": 0.6843220347965844,
+      "learning_rate": 2.573624997138565e-05,
+      "loss": 1.0134,
+      "step": 135
+    },
+    {
+      "epoch": 0.28874734607218683,
+      "grad_norm": 0.6783409077200856,
+      "learning_rate": 2.5662362869906123e-05,
+      "loss": 0.9991,
+      "step": 136
+    },
+    {
+      "epoch": 0.2908704883227176,
+      "grad_norm": 0.6082410802648054,
+      "learning_rate": 2.558794910110761e-05,
+      "loss": 0.9365,
+      "step": 137
+    },
+    {
+      "epoch": 0.2929936305732484,
+      "grad_norm": 0.6110004204749758,
+      "learning_rate": 2.5513012340657557e-05,
+      "loss": 0.9458,
+      "step": 138
+    },
+    {
+      "epoch": 0.2951167728237792,
+      "grad_norm": 0.6537047244906001,
+      "learning_rate": 2.5437556290056575e-05,
+      "loss": 0.9348,
+      "step": 139
+    },
+    {
+      "epoch": 0.29723991507430997,
+      "grad_norm": 0.6237035022111638,
+      "learning_rate": 2.5361584676455627e-05,
+      "loss": 0.9963,
+      "step": 140
+    },
+    {
+      "epoch": 0.29936305732484075,
+      "grad_norm": 0.658465414967936,
+      "learning_rate": 2.5285101252471874e-05,
+      "loss": 0.9365,
+      "step": 141
+    },
+    {
+      "epoch": 0.30148619957537154,
+      "grad_norm": 0.6275241420274199,
+      "learning_rate": 2.5208109796003364e-05,
+      "loss": 1.0091,
+      "step": 142
+    },
+    {
+      "epoch": 0.3036093418259023,
+      "grad_norm": 0.6338044548670078,
+      "learning_rate": 2.5130614110042415e-05,
+      "loss": 1.0313,
+      "step": 143
+    },
+    {
+      "epoch": 0.3057324840764331,
+      "grad_norm": 0.6708789723264734,
+      "learning_rate": 2.5052618022487733e-05,
+      "loss": 0.9077,
+      "step": 144
+    },
+    {
+      "epoch": 0.3078556263269639,
+      "grad_norm": 0.6673243086867044,
+      "learning_rate": 2.4974125385955374e-05,
+      "loss": 0.9733,
+      "step": 145
+    },
+    {
+      "epoch": 0.3099787685774947,
+      "grad_norm": 0.6708310719412189,
+      "learning_rate": 2.4895140077588412e-05,
+      "loss": 0.9224,
+      "step": 146
+    },
+    {
+      "epoch": 0.31210191082802546,
+      "grad_norm": 0.744990991437575,
+      "learning_rate": 2.481566599886546e-05,
+      "loss": 0.9491,
+      "step": 147
+    },
+    {
+      "epoch": 0.31422505307855625,
+      "grad_norm": 0.6620719317323561,
+      "learning_rate": 2.473570707540793e-05,
+      "loss": 1.0361,
+      "step": 148
+    },
+    {
+      "epoch": 0.31634819532908703,
+      "grad_norm": 0.6265774077983216,
+      "learning_rate": 2.4655267256786126e-05,
+      "loss": 1.0269,
+      "step": 149
+    },
+    {
+      "epoch": 0.3184713375796178,
+      "grad_norm": 0.6861510574862938,
+      "learning_rate": 2.4574350516324176e-05,
+      "loss": 0.9546,
+      "step": 150
+    },
+    {
+      "epoch": 0.3205944798301486,
+      "grad_norm": 0.6074369745423525,
+      "learning_rate": 2.4492960850903757e-05,
+      "loss": 0.9913,
+      "step": 151
+    },
+    {
+      "epoch": 0.3227176220806794,
+      "grad_norm": 0.6170833154929086,
+      "learning_rate": 2.4411102280766658e-05,
+      "loss": 0.9914,
+      "step": 152
+    },
+    {
+      "epoch": 0.3248407643312102,
+      "grad_norm": 0.6453490886193386,
+      "learning_rate": 2.432877884931623e-05,
+      "loss": 1.0185,
+      "step": 153
+    },
+    {
+      "epoch": 0.32696390658174096,
+      "grad_norm": 0.6376312745074503,
+      "learning_rate": 2.4245994622917636e-05,
+      "loss": 1.014,
+      "step": 154
+    },
+    {
+      "epoch": 0.32908704883227174,
+      "grad_norm": 0.5742876859098425,
+      "learning_rate": 2.4162753690696998e-05,
+      "loss": 1.0054,
+      "step": 155
+    },
+    {
+      "epoch": 0.33121019108280253,
+      "grad_norm": 0.6636167519331397,
+      "learning_rate": 2.4079060164339427e-05,
+      "loss": 0.988,
+      "step": 156
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.6393378453153665,
+      "learning_rate": 2.3994918177885906e-05,
+      "loss": 1.0362,
+      "step": 157
+    },
+    {
+      "epoch": 0.3354564755838641,
+      "grad_norm": 0.5723846211227461,
+      "learning_rate": 2.391033188752911e-05,
+      "loss": 0.9425,
+      "step": 158
+    },
+    {
+      "epoch": 0.3375796178343949,
+      "grad_norm": 0.6701423527493513,
+      "learning_rate": 2.38253054714081e-05,
+      "loss": 1.0224,
+      "step": 159
+    },
+    {
+      "epoch": 0.33970276008492567,
+      "grad_norm": 0.6385278318499282,
+      "learning_rate": 2.3739843129401932e-05,
+      "loss": 0.9562,
+      "step": 160
+    },
+    {
+      "epoch": 0.34182590233545646,
+      "grad_norm": 0.6587017894882612,
+      "learning_rate": 2.365394908292224e-05,
+      "loss": 0.9237,
+      "step": 161
+    },
+    {
+      "epoch": 0.34394904458598724,
+      "grad_norm": 0.6205619011806367,
+      "learning_rate": 2.356762757470468e-05,
+      "loss": 0.9645,
+      "step": 162
+    },
+    {
+      "epoch": 0.346072186836518,
+      "grad_norm": 0.6388406844797557,
+      "learning_rate": 2.3480882868599383e-05,
+      "loss": 0.9731,
+      "step": 163
+    },
+    {
+      "epoch": 0.3481953290870488,
+      "grad_norm": 0.6390704004642666,
+      "learning_rate": 2.3393719249360335e-05,
+      "loss": 1.0109,
+      "step": 164
+    },
+    {
+      "epoch": 0.3503184713375796,
+      "grad_norm": 0.6134473572194278,
+      "learning_rate": 2.3306141022433728e-05,
+      "loss": 1.0178,
+      "step": 165
+    },
+    {
+      "epoch": 0.3524416135881104,
+      "grad_norm": 0.6031953979375974,
+      "learning_rate": 2.3218152513745306e-05,
+      "loss": 0.9301,
+      "step": 166
+    },
+    {
+      "epoch": 0.35456475583864117,
+      "grad_norm": 0.6766275451516183,
+      "learning_rate": 2.3129758069486665e-05,
+      "loss": 0.9615,
+      "step": 167
+    },
+    {
+      "epoch": 0.35668789808917195,
+      "grad_norm": 0.5864063100045787,
+      "learning_rate": 2.3040962055900598e-05,
+      "loss": 1.0087,
+      "step": 168
+    },
+    {
+      "epoch": 0.35881104033970274,
+      "grad_norm": 0.6410164877808251,
+      "learning_rate": 2.2951768859065405e-05,
+      "loss": 0.9828,
+      "step": 169
+    },
+    {
+      "epoch": 0.3609341825902335,
+      "grad_norm": 0.6182510793860089,
+      "learning_rate": 2.2862182884678243e-05,
+      "loss": 0.914,
+      "step": 170
+    },
+    {
+      "epoch": 0.3630573248407643,
+      "grad_norm": 0.6109084937070998,
+      "learning_rate": 2.2772208557837526e-05,
+      "loss": 0.9526,
+      "step": 171
+    },
+    {
+      "epoch": 0.3651804670912951,
+      "grad_norm": 0.6117550209801461,
+      "learning_rate": 2.2681850322824324e-05,
+      "loss": 0.9419,
+      "step": 172
+    },
+    {
+      "epoch": 0.3673036093418259,
+      "grad_norm": 0.6057950347730594,
+      "learning_rate": 2.2591112642882853e-05,
+      "loss": 0.9378,
+      "step": 173
+    },
+    {
+      "epoch": 0.36942675159235666,
+      "grad_norm": 0.6090436491314833,
+      "learning_rate": 2.25e-05,
+      "loss": 0.9761,
+      "step": 174
+    },
+    {
+      "epoch": 0.37154989384288745,
+      "grad_norm": 0.6192774184724117,
+      "learning_rate": 2.2408516894683952e-05,
+      "loss": 0.9504,
+      "step": 175
+    },
+    {
+      "epoch": 0.37367303609341823,
+      "grad_norm": 0.6513174814734923,
+      "learning_rate": 2.2316667845741885e-05,
+      "loss": 0.9883,
+      "step": 176
+    },
+    {
+      "epoch": 0.37579617834394907,
+      "grad_norm": 0.6560644930725986,
+      "learning_rate": 2.2224457390056747e-05,
+      "loss": 0.9457,
+      "step": 177
+    },
+    {
+      "epoch": 0.37791932059447986,
+      "grad_norm": 0.6116671564494237,
+      "learning_rate": 2.2131890082363176e-05,
+      "loss": 0.9849,
+      "step": 178
+    },
+    {
+      "epoch": 0.38004246284501064,
+      "grad_norm": 0.6596643709604704,
+      "learning_rate": 2.2038970495022515e-05,
+      "loss": 0.9239,
+      "step": 179
+    },
+    {
+      "epoch": 0.3821656050955414,
+      "grad_norm": 0.6902003864146642,
+      "learning_rate": 2.194570321779695e-05,
+      "loss": 1.0142,
+      "step": 180
+    },
+    {
+      "epoch": 0.3842887473460722,
+      "grad_norm": 0.6472331908547782,
+      "learning_rate": 2.185209285762281e-05,
+      "loss": 0.9807,
+      "step": 181
+    },
+    {
+      "epoch": 0.386411889596603,
+      "grad_norm": 0.5879920270887264,
+      "learning_rate": 2.1758144038383e-05,
+      "loss": 0.9983,
+      "step": 182
+    },
+    {
+      "epoch": 0.3885350318471338,
+      "grad_norm": 0.647220005434279,
+      "learning_rate": 2.166386140067861e-05,
+      "loss": 0.9361,
+      "step": 183
+    },
+    {
+      "epoch": 0.39065817409766457,
+      "grad_norm": 0.6422591531038985,
+      "learning_rate": 2.1569249601599697e-05,
+      "loss": 0.9166,
+      "step": 184
+    },
+    {
+      "epoch": 0.39278131634819535,
+      "grad_norm": 0.6068003709339048,
+      "learning_rate": 2.147431331449522e-05,
+      "loss": 0.987,
+      "step": 185
+    },
+    {
+      "epoch": 0.39490445859872614,
+      "grad_norm": 0.7073279288261982,
+      "learning_rate": 2.1379057228742246e-05,
+      "loss": 1.0215,
+      "step": 186
+    },
+    {
+      "epoch": 0.3970276008492569,
+      "grad_norm": 0.6123796585365825,
+      "learning_rate": 2.128348604951428e-05,
+      "loss": 1.0122,
+      "step": 187
+    },
+    {
+      "epoch": 0.3991507430997877,
+      "grad_norm": 0.5646703919435287,
+      "learning_rate": 2.1187604497548862e-05,
+      "loss": 1.0163,
+      "step": 188
+    },
+    {
+      "epoch": 0.4012738853503185,
+      "grad_norm": 0.707419026365204,
+      "learning_rate": 2.1091417308914406e-05,
+      "loss": 0.9333,
+      "step": 189
+    },
+    {
+      "epoch": 0.4033970276008493,
+      "grad_norm": 0.55876064307062,
+      "learning_rate": 2.0994929234776232e-05,
+      "loss": 0.8764,
+      "step": 190
+    },
+    {
+      "epoch": 0.40552016985138006,
+      "grad_norm": 0.6189008896390565,
+      "learning_rate": 2.089814504116191e-05,
+      "loss": 0.9441,
+      "step": 191
+    },
+    {
+      "epoch": 0.40764331210191085,
+      "grad_norm": 0.6487599142290061,
+      "learning_rate": 2.0801069508725815e-05,
+      "loss": 0.9467,
+      "step": 192
+    },
+    {
+      "epoch": 0.40976645435244163,
+      "grad_norm": 0.5800499138520817,
+      "learning_rate": 2.0703707432513006e-05,
+      "loss": 0.9793,
+      "step": 193
+    },
+    {
+      "epoch": 0.4118895966029724,
+      "grad_norm": 0.6260444366432476,
+      "learning_rate": 2.0606063621722374e-05,
+      "loss": 0.9631,
+      "step": 194
+    },
+    {
+      "epoch": 0.4140127388535032,
+      "grad_norm": 0.6434863222057013,
+      "learning_rate": 2.050814289946908e-05,
+      "loss": 0.9217,
+      "step": 195
+    },
+    {
+      "epoch": 0.416135881104034,
+      "grad_norm": 0.6279442230026151,
+      "learning_rate": 2.0409950102546334e-05,
+      "loss": 0.9156,
+      "step": 196
+    },
+    {
+      "epoch": 0.4182590233545648,
+      "grad_norm": 0.6004879407537032,
+      "learning_rate": 2.0311490081186458e-05,
+      "loss": 0.9706,
+      "step": 197
+    },
+    {
+      "epoch": 0.42038216560509556,
+      "grad_norm": 0.6527914761227536,
+      "learning_rate": 2.021276769882133e-05,
+      "loss": 1.0022,
+      "step": 198
+    },
+    {
+      "epoch": 0.42250530785562634,
+      "grad_norm": 0.6666071133368344,
+      "learning_rate": 2.0113787831842155e-05,
+      "loss": 0.938,
+      "step": 199
+    },
+    {
+      "epoch": 0.42462845010615713,
+      "grad_norm": 0.613300047105262,
+      "learning_rate": 2.0014555369358576e-05,
+      "loss": 0.9555,
+      "step": 200
+    },
+    {
+      "epoch": 0.4267515923566879,
+      "grad_norm": 0.6329025879345046,
+      "learning_rate": 1.9915075212957192e-05,
+      "loss": 0.974,
+      "step": 201
+    },
+    {
+      "epoch": 0.4288747346072187,
+      "grad_norm": 0.6460699216963086,
+      "learning_rate": 1.9815352276459447e-05,
+      "loss": 0.9687,
+      "step": 202
+    },
+    {
+      "epoch": 0.4309978768577495,
+      "grad_norm": 0.5693282535126375,
+      "learning_rate": 1.9715391485678893e-05,
+      "loss": 0.9225,
+      "step": 203
+    },
+    {
+      "epoch": 0.43312101910828027,
+      "grad_norm": 0.6028040675891717,
+      "learning_rate": 1.961519777817791e-05,
+      "loss": 0.9561,
+      "step": 204
+    },
+    {
+      "epoch": 0.43524416135881105,
+      "grad_norm": 0.5964782950749927,
+      "learning_rate": 1.9514776103023783e-05,
+      "loss": 1.0385,
+      "step": 205
+    },
+    {
+      "epoch": 0.43736730360934184,
+      "grad_norm": 0.6258614863873178,
+      "learning_rate": 1.9414131420544258e-05,
+      "loss": 0.9441,
+      "step": 206
+    },
+    {
+      "epoch": 0.4394904458598726,
+      "grad_norm": 0.5843981439628442,
+      "learning_rate": 1.9313268702082536e-05,
+      "loss": 0.9485,
+      "step": 207
+    },
+    {
+      "epoch": 0.4416135881104034,
+      "grad_norm": 0.6272278606324312,
+      "learning_rate": 1.921219292975169e-05,
+      "loss": 0.911,
+      "step": 208
+    },
+    {
+      "epoch": 0.4437367303609342,
+      "grad_norm": 0.571371168999436,
+      "learning_rate": 1.9110909096188606e-05,
+      "loss": 0.9161,
+      "step": 209
+    },
+    {
+      "epoch": 0.445859872611465,
+      "grad_norm": 0.6219809311888109,
+      "learning_rate": 1.900942220430733e-05,
+      "loss": 0.9759,
+      "step": 210
+    },
+    {
+      "epoch": 0.44798301486199577,
+      "grad_norm": 0.5924793719813729,
+      "learning_rate": 1.890773726705198e-05,
+      "loss": 0.9515,
+      "step": 211
+    },
+    {
+      "epoch": 0.45010615711252655,
+      "grad_norm": 0.6083549766898838,
+      "learning_rate": 1.880585930714914e-05,
+      "loss": 0.9277,
+      "step": 212
+    },
+    {
+      "epoch": 0.45222929936305734,
+      "grad_norm": 0.6382145126359631,
+      "learning_rate": 1.8703793356859717e-05,
+      "loss": 0.9268,
+      "step": 213
+    },
+    {
+      "epoch": 0.4543524416135881,
+      "grad_norm": 0.5977568426364863,
+      "learning_rate": 1.8601544457730426e-05,
+      "loss": 0.9081,
+      "step": 214
+    },
+    {
+      "epoch": 0.4564755838641189,
+      "grad_norm": 0.5970654264593069,
+      "learning_rate": 1.849911766034472e-05,
+      "loss": 0.9194,
+      "step": 215
+    },
+    {
+      "epoch": 0.4585987261146497,
+      "grad_norm": 0.6596541084844906,
+      "learning_rate": 1.8396518024073356e-05,
+      "loss": 0.9674,
+      "step": 216
+    },
+    {
+      "epoch": 0.4607218683651805,
+      "grad_norm": 0.6165089515789794,
+      "learning_rate": 1.8293750616824443e-05,
+      "loss": 1.0102,
+      "step": 217
+    },
+    {
+      "epoch": 0.46284501061571126,
+      "grad_norm": 0.5710130818893961,
+      "learning_rate": 1.819082051479315e-05,
+      "loss": 0.9414,
+      "step": 218
+    },
+    {
+      "epoch": 0.46496815286624205,
+      "grad_norm": 0.5709493042606094,
+      "learning_rate": 1.8087732802210956e-05,
+      "loss": 0.9246,
+      "step": 219
+    },
+    {
+      "epoch": 0.46709129511677283,
+      "grad_norm": 0.5713801994508075,
+      "learning_rate": 1.7984492571094497e-05,
+      "loss": 0.9442,
+      "step": 220
+    },
+    {
+      "epoch": 0.4692144373673036,
+      "grad_norm": 0.6212813583103608,
+      "learning_rate": 1.788110492099407e-05,
+      "loss": 1.0118,
+      "step": 221
+    },
+    {
+      "epoch": 0.4713375796178344,
+      "grad_norm": 0.5542674209590273,
+      "learning_rate": 1.777757495874174e-05,
+      "loss": 0.8859,
+      "step": 222
+    },
+    {
+      "epoch": 0.4734607218683652,
+      "grad_norm": 0.6231448367618974,
+      "learning_rate": 1.7673907798199054e-05,
+      "loss": 0.9212,
+      "step": 223
+    },
+    {
+      "epoch": 0.47558386411889597,
+      "grad_norm": 0.5788006380629036,
+      "learning_rate": 1.75701085600045e-05,
+      "loss": 0.9956,
+      "step": 224
+    },
+    {
+      "epoch": 0.47770700636942676,
+      "grad_norm": 0.667025102007472,
+      "learning_rate": 1.7466182371320518e-05,
+      "loss": 0.9258,
+      "step": 225
+    },
+    {
+      "epoch": 0.47983014861995754,
+      "grad_norm": 0.5789996963104334,
+      "learning_rate": 1.7362134365580268e-05,
+      "loss": 0.922,
+      "step": 226
+    },
+    {
+      "epoch": 0.4819532908704883,
+      "grad_norm": 0.5380156350473858,
+      "learning_rate": 1.7257969682234073e-05,
+      "loss": 0.9583,
+      "step": 227
+    },
+    {
+      "epoch": 0.4840764331210191,
+      "grad_norm": 0.5920365256251041,
+      "learning_rate": 1.7153693466495538e-05,
+      "loss": 0.9588,
+      "step": 228
+    },
+    {
+      "epoch": 0.4861995753715499,
+      "grad_norm": 0.6254871880111138,
+      "learning_rate": 1.7049310869087402e-05,
+      "loss": 0.9793,
+      "step": 229
+    },
+    {
+      "epoch": 0.4883227176220807,
+      "grad_norm": 0.5740149209669835,
+      "learning_rate": 1.6944827045987148e-05,
+      "loss": 0.9483,
+      "step": 230
+    },
+    {
+      "epoch": 0.49044585987261147,
+      "grad_norm": 0.6132558502364226,
+      "learning_rate": 1.6840247158172277e-05,
+      "loss": 0.9082,
+      "step": 231
+    },
+    {
+      "epoch": 0.49256900212314225,
+      "grad_norm": 0.6259226448543737,
+      "learning_rate": 1.673557637136542e-05,
+      "loss": 1.0396,
+      "step": 232
+    },
+    {
+      "epoch": 0.49469214437367304,
+      "grad_norm": 1.0920922511810338,
+      "learning_rate": 1.663081985577916e-05,
+      "loss": 0.9112,
+      "step": 233
+    },
+    {
+      "epoch": 0.4968152866242038,
+      "grad_norm": 0.5773074384867334,
+      "learning_rate": 1.652598278586065e-05,
+      "loss": 0.8965,
+      "step": 234
+    },
+    {
+      "epoch": 0.4989384288747346,
+      "grad_norm": 0.6309124283703914,
+      "learning_rate": 1.6421070340036026e-05,
+      "loss": 0.9624,
+      "step": 235
+    },
+    {
+      "epoch": 0.5010615711252654,
+      "grad_norm": 0.5941173912410761,
+      "learning_rate": 1.631608770045461e-05,
+      "loss": 0.8636,
+      "step": 236
+    },
+    {
+      "epoch": 0.5031847133757962,
+      "grad_norm": 0.6981644261413696,
+      "learning_rate": 1.6211040052732958e-05,
+      "loss": 0.9533,
+      "step": 237
+    },
+    {
+      "epoch": 0.505307855626327,
+      "grad_norm": 0.5925395527066952,
+      "learning_rate": 1.6105932585698687e-05,
+      "loss": 0.9811,
+      "step": 238
+    },
+    {
+      "epoch": 0.5074309978768577,
+      "grad_norm": 0.5889941819588222,
+      "learning_rate": 1.6000770491134195e-05,
+      "loss": 0.9276,
+      "step": 239
+    },
+    {
+      "epoch": 0.5095541401273885,
+      "grad_norm": 0.6054802307465149,
+      "learning_rate": 1.5895558963520207e-05,
+      "loss": 1.0013,
+      "step": 240
+    },
+    {
+      "epoch": 0.5116772823779193,
+      "grad_norm": 0.5854921389762675,
+      "learning_rate": 1.5790303199779194e-05,
+      "loss": 0.938,
+      "step": 241
+    },
+    {
+      "epoch": 0.5138004246284501,
+      "grad_norm": 0.6156629079015952,
+      "learning_rate": 1.568500839901867e-05,
+      "loss": 0.8998,
+      "step": 242
+    },
+    {
+      "epoch": 0.5159235668789809,
+      "grad_norm": 0.5942832812673114,
+      "learning_rate": 1.5579679762274377e-05,
+      "loss": 0.906,
+      "step": 243
+    },
+    {
+      "epoch": 0.5180467091295117,
+      "grad_norm": 0.6245210048609385,
+      "learning_rate": 1.5474322492253382e-05,
+      "loss": 0.9881,
+      "step": 244
+    },
+    {
+      "epoch": 0.5201698513800425,
+      "grad_norm": 0.5813078674079436,
+      "learning_rate": 1.5368941793077115e-05,
+      "loss": 0.9948,
+      "step": 245
+    },
+    {
+      "epoch": 0.5222929936305732,
+      "grad_norm": 0.5729126067790324,
+      "learning_rate": 1.5263542870024263e-05,
+      "loss": 0.9782,
+      "step": 246
+    },
+    {
+      "epoch": 0.524416135881104,
+      "grad_norm": 0.6115976256801431,
+      "learning_rate": 1.5158130929273695e-05,
+      "loss": 0.9586,
+      "step": 247
+    },
+    {
+      "epoch": 0.5265392781316348,
+      "grad_norm": 0.5677823931669936,
+      "learning_rate": 1.505271117764728e-05,
+      "loss": 0.9273,
+      "step": 248
+    },
+    {
+      "epoch": 0.5286624203821656,
+      "grad_norm": 0.6337456306602008,
+      "learning_rate": 1.494728882235272e-05,
+      "loss": 0.9653,
+      "step": 249
+    },
+    {
+      "epoch": 0.5307855626326964,
+      "grad_norm": 0.5908641833311927,
+      "learning_rate": 1.4841869070726307e-05,
+      "loss": 0.9783,
+      "step": 250
+    },
+    {
+      "epoch": 0.5329087048832272,
+      "grad_norm": 0.6315479071635102,
+      "learning_rate": 1.473645712997574e-05,
+      "loss": 0.9341,
+      "step": 251
+    },
+    {
+      "epoch": 0.535031847133758,
+      "grad_norm": 0.6461862581364697,
+      "learning_rate": 1.4631058206922886e-05,
+      "loss": 0.9351,
+      "step": 252
+    },
+    {
+      "epoch": 0.5371549893842887,
+      "grad_norm": 0.6114227562639025,
+      "learning_rate": 1.4525677507746615e-05,
+      "loss": 0.9614,
+      "step": 253
+    },
+    {
+      "epoch": 0.5392781316348195,
+      "grad_norm": 0.5821157750989183,
+      "learning_rate": 1.4420320237725632e-05,
+      "loss": 1.0022,
+      "step": 254
+    },
+    {
+      "epoch": 0.5414012738853503,
+      "grad_norm": 0.6176480452727935,
+      "learning_rate": 1.4314991600981336e-05,
+      "loss": 0.9242,
+      "step": 255
+    },
+    {
+      "epoch": 0.5435244161358811,
+      "grad_norm": 0.591935718739984,
+      "learning_rate": 1.4209696800220807e-05,
+      "loss": 0.9189,
+      "step": 256
+    },
+    {
+      "epoch": 0.5456475583864119,
+      "grad_norm": 0.6355341920694051,
+      "learning_rate": 1.4104441036479797e-05,
+      "loss": 0.9276,
+      "step": 257
+    },
+    {
+      "epoch": 0.5477707006369427,
+      "grad_norm": 0.6077185944726149,
+      "learning_rate": 1.399922950886581e-05,
+      "loss": 0.9697,
+      "step": 258
+    },
+    {
+      "epoch": 0.5498938428874734,
+      "grad_norm": 0.5590992275595242,
+      "learning_rate": 1.3894067414301315e-05,
+      "loss": 0.9377,
+      "step": 259
+    },
+    {
+      "epoch": 0.5520169851380042,
+      "grad_norm": 0.5998950581637562,
+      "learning_rate": 1.3788959947267042e-05,
+      "loss": 0.909,
+      "step": 260
+    },
+    {
+      "epoch": 0.554140127388535,
+      "grad_norm": 0.6965397105135438,
+      "learning_rate": 1.3683912299545391e-05,
+      "loss": 0.9523,
+      "step": 261
+    },
+    {
+      "epoch": 0.5562632696390658,
+      "grad_norm": 0.635211817521203,
+      "learning_rate": 1.3578929659963977e-05,
+      "loss": 0.9972,
+      "step": 262
+    },
+    {
+      "epoch": 0.5583864118895966,
+      "grad_norm": 0.5385969763281576,
+      "learning_rate": 1.347401721413935e-05,
+      "loss": 0.8788,
+      "step": 263
+    },
+    {
+      "epoch": 0.5605095541401274,
+      "grad_norm": 0.5747284767597718,
+      "learning_rate": 1.336918014422084e-05,
+      "loss": 0.9345,
+      "step": 264
+    },
+    {
+      "epoch": 0.5626326963906582,
+      "grad_norm": 0.6397363727759597,
+      "learning_rate": 1.3264423628634583e-05,
+      "loss": 0.9852,
+      "step": 265
+    },
+    {
+      "epoch": 0.564755838641189,
+      "grad_norm": 0.6401143379139339,
+      "learning_rate": 1.3159752841827724e-05,
+      "loss": 0.9803,
+      "step": 266
+    },
+    {
+      "epoch": 0.5668789808917197,
+      "grad_norm": 0.5704645508381336,
+      "learning_rate": 1.3055172954012856e-05,
+      "loss": 0.9394,
+      "step": 267
+    },
+    {
+      "epoch": 0.5690021231422505,
+      "grad_norm": 0.5783312592479926,
+      "learning_rate": 1.2950689130912599e-05,
+      "loss": 1.0313,
+      "step": 268
+    },
+    {
+      "epoch": 0.5711252653927813,
+      "grad_norm": 0.5472325263398086,
+      "learning_rate": 1.2846306533504465e-05,
+      "loss": 0.9307,
+      "step": 269
+    },
+    {
+      "epoch": 0.5732484076433121,
+      "grad_norm": 0.5949754852554261,
+      "learning_rate": 1.2742030317765933e-05,
+      "loss": 0.9463,
+      "step": 270
+    },
+    {
+      "epoch": 0.5753715498938429,
+      "grad_norm": 0.5642237939718832,
+      "learning_rate": 1.2637865634419735e-05,
+      "loss": 0.9339,
+      "step": 271
+    },
+    {
+      "epoch": 0.5774946921443737,
+      "grad_norm": 0.5950396421063334,
+      "learning_rate": 1.2533817628679485e-05,
+      "loss": 0.9871,
+      "step": 272
+    },
+    {
+      "epoch": 0.5796178343949044,
+      "grad_norm": 0.555308435607654,
+      "learning_rate": 1.24298914399955e-05,
+      "loss": 0.9088,
+      "step": 273
+    },
+    {
+      "epoch": 0.5817409766454352,
+      "grad_norm": 0.5543717603232199,
+      "learning_rate": 1.2326092201800948e-05,
+      "loss": 0.9858,
+      "step": 274
+    },
+    {
+      "epoch": 0.583864118895966,
+      "grad_norm": 0.5768748399190285,
+      "learning_rate": 1.2222425041258267e-05,
+      "loss": 0.9933,
+      "step": 275
+    },
+    {
+      "epoch": 0.5859872611464968,
+      "grad_norm": 0.6144146862154265,
+      "learning_rate": 1.2118895079005929e-05,
+      "loss": 0.966,
+      "step": 276
+    },
+    {
+      "epoch": 0.5881104033970276,
+      "grad_norm": 0.6371726541609838,
+      "learning_rate": 1.2015507428905509e-05,
+      "loss": 0.9366,
+      "step": 277
+    },
+    {
+      "epoch": 0.5902335456475584,
+      "grad_norm": 0.5757938188544394,
+      "learning_rate": 1.1912267197789047e-05,
+      "loss": 0.9364,
+      "step": 278
+    },
+    {
+      "epoch": 0.5923566878980892,
+      "grad_norm": 0.6158043621368744,
+      "learning_rate": 1.1809179485206847e-05,
+      "loss": 0.9741,
+      "step": 279
+    },
+    {
+      "epoch": 0.5944798301486199,
+      "grad_norm": 0.5414978480240109,
+      "learning_rate": 1.1706249383175558e-05,
+      "loss": 0.9869,
+      "step": 280
+    },
+    {
+      "epoch": 0.5966029723991507,
+      "grad_norm": 0.6152469998780976,
+      "learning_rate": 1.1603481975926643e-05,
+      "loss": 0.9781,
+      "step": 281
+    },
+    {
+      "epoch": 0.5987261146496815,
+      "grad_norm": 0.6082547306754248,
+      "learning_rate": 1.1500882339655278e-05,
+      "loss": 0.9069,
+      "step": 282
+    },
+    {
+      "epoch": 0.6008492569002123,
+      "grad_norm": 0.6137321132683284,
+      "learning_rate": 1.1398455542269578e-05,
+      "loss": 0.9701,
+      "step": 283
+    },
+    {
+      "epoch": 0.6029723991507431,
+      "grad_norm": 0.5745752002120236,
+      "learning_rate": 1.1296206643140284e-05,
+      "loss": 0.9598,
+      "step": 284
+    },
+    {
+      "epoch": 0.6050955414012739,
+      "grad_norm": 0.7041392825748041,
+      "learning_rate": 1.1194140692850863e-05,
+      "loss": 0.979,
+      "step": 285
+    },
+    {
+      "epoch": 0.6072186836518046,
+      "grad_norm": 0.5744290113136731,
+      "learning_rate": 1.1092262732948017e-05,
+      "loss": 0.9193,
+      "step": 286
+    },
+    {
+      "epoch": 0.6093418259023354,
+      "grad_norm": 0.5425442729898244,
+      "learning_rate": 1.0990577795692676e-05,
+      "loss": 0.9141,
+      "step": 287
+    },
+    {
+      "epoch": 0.6114649681528662,
+      "grad_norm": 0.5687470133452739,
+      "learning_rate": 1.0889090903811397e-05,
+      "loss": 0.958,
+      "step": 288
+    },
+    {
+      "epoch": 0.613588110403397,
+      "grad_norm": 0.567846934811383,
+      "learning_rate": 1.0787807070248306e-05,
+      "loss": 0.8882,
+      "step": 289
+    },
+    {
+      "epoch": 0.6157112526539278,
+      "grad_norm": 0.5646293628586256,
+      "learning_rate": 1.068673129791747e-05,
+      "loss": 0.9476,
+      "step": 290
+    },
+    {
+      "epoch": 0.6178343949044586,
+      "grad_norm": 0.5568078722816252,
+      "learning_rate": 1.0585868579455745e-05,
+      "loss": 0.9549,
+      "step": 291
+    },
+    {
+      "epoch": 0.6199575371549894,
+      "grad_norm": 0.5564148135548681,
+      "learning_rate": 1.0485223896976221e-05,
+      "loss": 0.9683,
+      "step": 292
+    },
+    {
+      "epoch": 0.6220806794055201,
+      "grad_norm": 0.5779558456789065,
+      "learning_rate": 1.0384802221822098e-05,
+      "loss": 0.981,
+      "step": 293
+    },
+    {
+      "epoch": 0.6242038216560509,
+      "grad_norm": 0.579183894606748,
+      "learning_rate": 1.028460851432111e-05,
+      "loss": 0.8776,
+      "step": 294
+    },
+    {
+      "epoch": 0.6263269639065817,
+      "grad_norm": 0.5562390759302435,
+      "learning_rate": 1.0184647723540557e-05,
+      "loss": 0.9366,
+      "step": 295
+    },
+    {
+      "epoch": 0.6284501061571125,
+      "grad_norm": 0.6039425393322856,
+      "learning_rate": 1.0084924787042809e-05,
+      "loss": 0.9764,
+      "step": 296
+    },
+    {
+      "epoch": 0.6305732484076433,
+      "grad_norm": 0.5990295929838292,
+      "learning_rate": 9.985444630641426e-06,
+      "loss": 0.9575,
+      "step": 297
+    },
+    {
+      "epoch": 0.6326963906581741,
+      "grad_norm": 0.5334970805071061,
+      "learning_rate": 9.886212168157848e-06,
+      "loss": 0.9255,
+      "step": 298
+    },
+    {
+      "epoch": 0.6348195329087049,
+      "grad_norm": 0.5458651078507479,
+      "learning_rate": 9.787232301178669e-06,
+      "loss": 0.9658,
+      "step": 299
+    },
+    {
+      "epoch": 0.6369426751592356,
+      "grad_norm": 0.570409644049972,
+      "learning_rate": 9.688509918813547e-06,
+      "loss": 0.9055,
+      "step": 300
+    },
+    {
+      "epoch": 0.6390658174097664,
+      "grad_norm": 0.559661538879096,
+      "learning_rate": 9.590049897453668e-06,
+      "loss": 0.9829,
+      "step": 301
+    },
+    {
+      "epoch": 0.6411889596602972,
+      "grad_norm": 0.5594976911463373,
+      "learning_rate": 9.491857100530919e-06,
+      "loss": 0.9484,
+      "step": 302
+    },
+    {
+      "epoch": 0.643312101910828,
+      "grad_norm": 0.5272576598614247,
+      "learning_rate": 9.393936378277632e-06,
+      "loss": 0.9366,
+      "step": 303
+    },
+    {
+      "epoch": 0.6454352441613588,
+      "grad_norm": 0.5932545624966695,
+      "learning_rate": 9.296292567486997e-06,
+      "loss": 0.9009,
+      "step": 304
+    },
+    {
+      "epoch": 0.6475583864118896,
+      "grad_norm": 0.564218312285332,
+      "learning_rate": 9.198930491274188e-06,
+      "loss": 0.922,
+      "step": 305
+    },
+    {
+      "epoch": 0.6496815286624203,
+      "grad_norm": 0.556568715801767,
+      "learning_rate": 9.10185495883809e-06,
+      "loss": 0.9486,
+      "step": 306
+    },
+    {
+      "epoch": 0.6518046709129511,
+      "grad_norm": 0.545963447620699,
+      "learning_rate": 9.005070765223768e-06,
+      "loss": 0.9501,
+      "step": 307
+    },
+    {
+      "epoch": 0.6539278131634819,
+      "grad_norm": 0.5361391538001072,
+      "learning_rate": 8.908582691085593e-06,
+      "loss": 0.9539,
+      "step": 308
+    },
+    {
+      "epoch": 0.6560509554140127,
+      "grad_norm": 0.5993729956874576,
+      "learning_rate": 8.812395502451139e-06,
+      "loss": 1.0198,
+      "step": 309
+    },
+    {
+      "epoch": 0.6581740976645435,
+      "grad_norm": 0.5804015733760072,
+      "learning_rate": 8.716513950485725e-06,
+      "loss": 0.9764,
+      "step": 310
+    },
+    {
+      "epoch": 0.6602972399150743,
+      "grad_norm": 0.6255942117075288,
+      "learning_rate": 8.620942771257755e-06,
+      "loss": 0.926,
+      "step": 311
+    },
+    {
+      "epoch": 0.6624203821656051,
+      "grad_norm": 0.5661900360842226,
+      "learning_rate": 8.525686685504781e-06,
+      "loss": 0.8951,
+      "step": 312
+    },
+    {
+      "epoch": 0.6645435244161358,
+      "grad_norm": 0.5897078527410758,
+      "learning_rate": 8.430750398400309e-06,
+      "loss": 0.9424,
+      "step": 313
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.5099303542607264,
+      "learning_rate": 8.336138599321391e-06,
+      "loss": 0.9596,
+      "step": 314
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 471,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 157,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 46655615139840.0,
+  "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 sha256:453c3547a807cb2e9fa04ac2d0de5d456f008037ec1ec846c16b39c6989ac4c3
+size 8017