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 +1,30 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "max_position_embeddings": 40960,
+  "max_window_layers": 28,
+  "model_type": "qwen3",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "temperature": 0.6,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "4.51.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f901517f53d779f23ec27256b5dbb92e17516d088839bd358eb638eef032825
+size 2384234968

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3328fc929524112d4419183228e9dd28e66539edf365150efb57f445565fb6bb
+size 4768663315

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7063580a565cb4ab0c1d36b25d817a35a16d1f21f4a993a9f25cdba6efadcb9d
+size 14645

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9304b56ea23778a459e13aef7a6622bd168edce0ae7ae180268b26d9a68521a3
+size 1383

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e2a9bd8ec47523090615c73bcf7d5b1878cb8e32e20f73737588d1084bf8a6e
+size 1465

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 sha256:79894f433c8379c083f0666155142ab72946b24d053f443d6118f87133f10134
+size 11422751

tokenizer_config.json

@@ -0,0 +1,240 @@
+{
+  "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 message.content is string 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.content is string %}\n        {%- set content = message.content %}\n    {%- else %}\n        {%- set content = '' %}\n    {%- endif %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is string %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in content %}\n                {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n                {%- set content = content.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        {{- 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|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,2017 @@
+{
+  "best_global_step": 14058,
+  "best_metric": 0.6273267865180969,
+  "best_model_checkpoint": "./mcqa_qwen3_letter_m3/checkpoint-14058",
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 14058,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007113640405477503,
+      "grad_norm": 20.266469955444336,
+      "learning_rate": 4.551920341394026e-07,
+      "loss": 2.1439,
+      "step": 50
+    },
+    {
+      "epoch": 0.014227280810955006,
+      "grad_norm": 21.066892623901367,
+      "learning_rate": 9.293504030346136e-07,
+      "loss": 1.4471,
+      "step": 100
+    },
+    {
+      "epoch": 0.02134092121643251,
+      "grad_norm": 18.918508529663086,
+      "learning_rate": 1.4035087719298246e-06,
+      "loss": 1.3067,
+      "step": 150
+    },
+    {
+      "epoch": 0.028454561621910013,
+      "grad_norm": 9.563344955444336,
+      "learning_rate": 1.8681839734471315e-06,
+      "loss": 1.2745,
+      "step": 200
+    },
+    {
+      "epoch": 0.035568202027387516,
+      "grad_norm": 20.408035278320312,
+      "learning_rate": 2.3423423423423424e-06,
+      "loss": 1.2323,
+      "step": 250
+    },
+    {
+      "epoch": 0.04268184243286502,
+      "grad_norm": 17.020856857299805,
+      "learning_rate": 2.8165007112375536e-06,
+      "loss": 1.2476,
+      "step": 300
+    },
+    {
+      "epoch": 0.04979548283834252,
+      "grad_norm": 10.942744255065918,
+      "learning_rate": 3.2906590801327647e-06,
+      "loss": 1.2084,
+      "step": 350
+    },
+    {
+      "epoch": 0.056909123243820026,
+      "grad_norm": 11.239660263061523,
+      "learning_rate": 3.7648174490279754e-06,
+      "loss": 1.1453,
+      "step": 400
+    },
+    {
+      "epoch": 0.06402276364929753,
+      "grad_norm": 8.421320915222168,
+      "learning_rate": 4.2389758179231865e-06,
+      "loss": 1.1257,
+      "step": 450
+    },
+    {
+      "epoch": 0.07113640405477503,
+      "grad_norm": 7.429409027099609,
+      "learning_rate": 4.713134186818398e-06,
+      "loss": 1.0896,
+      "step": 500
+    },
+    {
+      "epoch": 0.07825004446025254,
+      "grad_norm": 8.3453950881958,
+      "learning_rate": 5.187292555713609e-06,
+      "loss": 1.0589,
+      "step": 550
+    },
+    {
+      "epoch": 0.08536368486573004,
+      "grad_norm": 9.157875061035156,
+      "learning_rate": 5.66145092460882e-06,
+      "loss": 1.0628,
+      "step": 600
+    },
+    {
+      "epoch": 0.09247732527120754,
+      "grad_norm": 9.339110374450684,
+      "learning_rate": 6.135609293504031e-06,
+      "loss": 1.0211,
+      "step": 650
+    },
+    {
+      "epoch": 0.09959096567668504,
+      "grad_norm": 7.010004043579102,
+      "learning_rate": 6.609767662399243e-06,
+      "loss": 1.0684,
+      "step": 700
+    },
+    {
+      "epoch": 0.10670460608216255,
+      "grad_norm": 8.618551254272461,
+      "learning_rate": 7.0839260312944525e-06,
+      "loss": 1.0467,
+      "step": 750
+    },
+    {
+      "epoch": 0.11381824648764005,
+      "grad_norm": 9.206182479858398,
+      "learning_rate": 7.558084400189663e-06,
+      "loss": 1.0855,
+      "step": 800
+    },
+    {
+      "epoch": 0.12093188689311755,
+      "grad_norm": 11.882657051086426,
+      "learning_rate": 8.032242769084875e-06,
+      "loss": 0.9897,
+      "step": 850
+    },
+    {
+      "epoch": 0.12804552729859506,
+      "grad_norm": 6.381322860717773,
+      "learning_rate": 8.506401137980086e-06,
+      "loss": 1.0227,
+      "step": 900
+    },
+    {
+      "epoch": 0.13515916770407255,
+      "grad_norm": 7.592447280883789,
+      "learning_rate": 8.980559506875298e-06,
+      "loss": 0.9508,
+      "step": 950
+    },
+    {
+      "epoch": 0.14227280810955006,
+      "grad_norm": 10.622550010681152,
+      "learning_rate": 9.454717875770508e-06,
+      "loss": 0.9916,
+      "step": 1000
+    },
+    {
+      "epoch": 0.14938644851502755,
+      "grad_norm": 16.836315155029297,
+      "learning_rate": 9.92887624466572e-06,
+      "loss": 0.9838,
+      "step": 1050
+    },
+    {
+      "epoch": 0.15650008892050507,
+      "grad_norm": 8.732917785644531,
+      "learning_rate": 1.040303461356093e-05,
+      "loss": 0.9678,
+      "step": 1100
+    },
+    {
+      "epoch": 0.16361372932598256,
+      "grad_norm": 10.75668716430664,
+      "learning_rate": 1.0877192982456142e-05,
+      "loss": 0.9535,
+      "step": 1150
+    },
+    {
+      "epoch": 0.17072736973146008,
+      "grad_norm": 10.604599952697754,
+      "learning_rate": 1.1351351351351352e-05,
+      "loss": 1.0164,
+      "step": 1200
+    },
+    {
+      "epoch": 0.17784101013693757,
+      "grad_norm": 11.192643165588379,
+      "learning_rate": 1.1825509720246564e-05,
+      "loss": 0.9811,
+      "step": 1250
+    },
+    {
+      "epoch": 0.18495465054241508,
+      "grad_norm": 11.667268753051758,
+      "learning_rate": 1.2299668089141774e-05,
+      "loss": 0.9329,
+      "step": 1300
+    },
+    {
+      "epoch": 0.19206829094789257,
+      "grad_norm": 12.122210502624512,
+      "learning_rate": 1.2773826458036987e-05,
+      "loss": 0.9693,
+      "step": 1350
+    },
+    {
+      "epoch": 0.1991819313533701,
+      "grad_norm": 10.696669578552246,
+      "learning_rate": 1.3247984826932197e-05,
+      "loss": 0.9118,
+      "step": 1400
+    },
+    {
+      "epoch": 0.20629557175884758,
+      "grad_norm": 9.4218168258667,
+      "learning_rate": 1.3722143195827407e-05,
+      "loss": 0.9242,
+      "step": 1450
+    },
+    {
+      "epoch": 0.2134092121643251,
+      "grad_norm": 15.12156867980957,
+      "learning_rate": 1.4196301564722618e-05,
+      "loss": 0.9161,
+      "step": 1500
+    },
+    {
+      "epoch": 0.22052285256980259,
+      "grad_norm": 10.341136932373047,
+      "learning_rate": 1.4670459933617828e-05,
+      "loss": 0.9112,
+      "step": 1550
+    },
+    {
+      "epoch": 0.2276364929752801,
+      "grad_norm": 8.621809005737305,
+      "learning_rate": 1.5144618302513041e-05,
+      "loss": 0.923,
+      "step": 1600
+    },
+    {
+      "epoch": 0.2347501333807576,
+      "grad_norm": 8.037166595458984,
+      "learning_rate": 1.561877667140825e-05,
+      "loss": 0.8646,
+      "step": 1650
+    },
+    {
+      "epoch": 0.2418637737862351,
+      "grad_norm": 12.160765647888184,
+      "learning_rate": 1.6092935040303463e-05,
+      "loss": 0.8811,
+      "step": 1700
+    },
+    {
+      "epoch": 0.2489774141917126,
+      "grad_norm": 9.850648880004883,
+      "learning_rate": 1.6567093409198673e-05,
+      "loss": 0.9158,
+      "step": 1750
+    },
+    {
+      "epoch": 0.2560910545971901,
+      "grad_norm": 6.849952220916748,
+      "learning_rate": 1.7041251778093886e-05,
+      "loss": 0.8903,
+      "step": 1800
+    },
+    {
+      "epoch": 0.26320469500266763,
+      "grad_norm": 7.7248334884643555,
+      "learning_rate": 1.7515410146989096e-05,
+      "loss": 0.8853,
+      "step": 1850
+    },
+    {
+      "epoch": 0.2703183354081451,
+      "grad_norm": 5.612997531890869,
+      "learning_rate": 1.7989568515884306e-05,
+      "loss": 0.8407,
+      "step": 1900
+    },
+    {
+      "epoch": 0.2774319758136226,
+      "grad_norm": 6.731159210205078,
+      "learning_rate": 1.846372688477952e-05,
+      "loss": 0.8447,
+      "step": 1950
+    },
+    {
+      "epoch": 0.28454561621910013,
+      "grad_norm": 8.783773422241211,
+      "learning_rate": 1.893788525367473e-05,
+      "loss": 0.8201,
+      "step": 2000
+    },
+    {
+      "epoch": 0.29165925662457765,
+      "grad_norm": 8.748090744018555,
+      "learning_rate": 1.9412043622569942e-05,
+      "loss": 0.8202,
+      "step": 2050
+    },
+    {
+      "epoch": 0.2987728970300551,
+      "grad_norm": 6.985095977783203,
+      "learning_rate": 1.9886201991465152e-05,
+      "loss": 0.8686,
+      "step": 2100
+    },
+    {
+      "epoch": 0.3058865374355326,
+      "grad_norm": 9.010405540466309,
+      "learning_rate": 1.9959947299077735e-05,
+      "loss": 0.8541,
+      "step": 2150
+    },
+    {
+      "epoch": 0.31300017784101014,
+      "grad_norm": 10.884967803955078,
+      "learning_rate": 1.9907246376811596e-05,
+      "loss": 0.9021,
+      "step": 2200
+    },
+    {
+      "epoch": 0.32011381824648766,
+      "grad_norm": 6.696805953979492,
+      "learning_rate": 1.9854545454545456e-05,
+      "loss": 0.8352,
+      "step": 2250
+    },
+    {
+      "epoch": 0.3272274586519651,
+      "grad_norm": 6.158838272094727,
+      "learning_rate": 1.9801844532279317e-05,
+      "loss": 0.8078,
+      "step": 2300
+    },
+    {
+      "epoch": 0.33434109905744264,
+      "grad_norm": 10.230331420898438,
+      "learning_rate": 1.9749143610013177e-05,
+      "loss": 0.8659,
+      "step": 2350
+    },
+    {
+      "epoch": 0.34145473946292015,
+      "grad_norm": 7.0724945068359375,
+      "learning_rate": 1.9696442687747038e-05,
+      "loss": 0.862,
+      "step": 2400
+    },
+    {
+      "epoch": 0.34856837986839767,
+      "grad_norm": 9.820209503173828,
+      "learning_rate": 1.9643741765480898e-05,
+      "loss": 0.8036,
+      "step": 2450
+    },
+    {
+      "epoch": 0.35568202027387513,
+      "grad_norm": 6.326645374298096,
+      "learning_rate": 1.959104084321476e-05,
+      "loss": 0.8255,
+      "step": 2500
+    },
+    {
+      "epoch": 0.36279566067935265,
+      "grad_norm": 4.881910800933838,
+      "learning_rate": 1.953939393939394e-05,
+      "loss": 0.861,
+      "step": 2550
+    },
+    {
+      "epoch": 0.36990930108483017,
+      "grad_norm": 6.977218151092529,
+      "learning_rate": 1.94866930171278e-05,
+      "loss": 0.8487,
+      "step": 2600
+    },
+    {
+      "epoch": 0.3770229414903077,
+      "grad_norm": 6.354854106903076,
+      "learning_rate": 1.943399209486166e-05,
+      "loss": 0.8122,
+      "step": 2650
+    },
+    {
+      "epoch": 0.38413658189578515,
+      "grad_norm": 6.679897308349609,
+      "learning_rate": 1.9381291172595522e-05,
+      "loss": 0.8239,
+      "step": 2700
+    },
+    {
+      "epoch": 0.39125022230126266,
+      "grad_norm": 7.958705902099609,
+      "learning_rate": 1.9328590250329382e-05,
+      "loss": 0.7808,
+      "step": 2750
+    },
+    {
+      "epoch": 0.3983638627067402,
+      "grad_norm": 6.450643062591553,
+      "learning_rate": 1.9275889328063243e-05,
+      "loss": 0.8359,
+      "step": 2800
+    },
+    {
+      "epoch": 0.4054775031122177,
+      "grad_norm": 5.986859321594238,
+      "learning_rate": 1.9223188405797103e-05,
+      "loss": 0.8278,
+      "step": 2850
+    },
+    {
+      "epoch": 0.41259114351769516,
+      "grad_norm": 8.40491771697998,
+      "learning_rate": 1.9170487483530964e-05,
+      "loss": 0.7932,
+      "step": 2900
+    },
+    {
+      "epoch": 0.4197047839231727,
+      "grad_norm": 7.491307735443115,
+      "learning_rate": 1.9117786561264824e-05,
+      "loss": 0.7932,
+      "step": 2950
+    },
+    {
+      "epoch": 0.4268184243286502,
+      "grad_norm": 5.475683689117432,
+      "learning_rate": 1.9065085638998684e-05,
+      "loss": 0.8292,
+      "step": 3000
+    },
+    {
+      "epoch": 0.4339320647341277,
+      "grad_norm": 6.489156723022461,
+      "learning_rate": 1.9012384716732545e-05,
+      "loss": 0.7585,
+      "step": 3050
+    },
+    {
+      "epoch": 0.44104570513960517,
+      "grad_norm": 6.417465686798096,
+      "learning_rate": 1.8959683794466402e-05,
+      "loss": 0.7628,
+      "step": 3100
+    },
+    {
+      "epoch": 0.4481593455450827,
+      "grad_norm": 4.215211868286133,
+      "learning_rate": 1.8906982872200266e-05,
+      "loss": 0.7418,
+      "step": 3150
+    },
+    {
+      "epoch": 0.4552729859505602,
+      "grad_norm": 5.503742218017578,
+      "learning_rate": 1.8854281949934126e-05,
+      "loss": 0.8346,
+      "step": 3200
+    },
+    {
+      "epoch": 0.4623866263560377,
+      "grad_norm": 5.3524651527404785,
+      "learning_rate": 1.8801581027667987e-05,
+      "loss": 0.8285,
+      "step": 3250
+    },
+    {
+      "epoch": 0.4695002667615152,
+      "grad_norm": 5.5221781730651855,
+      "learning_rate": 1.8748880105401844e-05,
+      "loss": 0.7363,
+      "step": 3300
+    },
+    {
+      "epoch": 0.4766139071669927,
+      "grad_norm": 7.9843268394470215,
+      "learning_rate": 1.8696179183135704e-05,
+      "loss": 0.7964,
+      "step": 3350
+    },
+    {
+      "epoch": 0.4837275475724702,
+      "grad_norm": 6.0546417236328125,
+      "learning_rate": 1.8643478260869568e-05,
+      "loss": 0.7696,
+      "step": 3400
+    },
+    {
+      "epoch": 0.49084118797794773,
+      "grad_norm": 5.521228313446045,
+      "learning_rate": 1.859077733860343e-05,
+      "loss": 0.7309,
+      "step": 3450
+    },
+    {
+      "epoch": 0.4979548283834252,
+      "grad_norm": 9.225383758544922,
+      "learning_rate": 1.853807641633729e-05,
+      "loss": 0.7869,
+      "step": 3500
+    },
+    {
+      "epoch": 0.5050684687889028,
+      "grad_norm": 4.0471601486206055,
+      "learning_rate": 1.8485375494071146e-05,
+      "loss": 0.7323,
+      "step": 3550
+    },
+    {
+      "epoch": 0.5121821091943802,
+      "grad_norm": 4.831805229187012,
+      "learning_rate": 1.8432674571805006e-05,
+      "loss": 0.7795,
+      "step": 3600
+    },
+    {
+      "epoch": 0.5192957495998577,
+      "grad_norm": 5.822267055511475,
+      "learning_rate": 1.837997364953887e-05,
+      "loss": 0.712,
+      "step": 3650
+    },
+    {
+      "epoch": 0.5264093900053353,
+      "grad_norm": 4.78722620010376,
+      "learning_rate": 1.832727272727273e-05,
+      "loss": 0.7408,
+      "step": 3700
+    },
+    {
+      "epoch": 0.5335230304108127,
+      "grad_norm": 5.776618480682373,
+      "learning_rate": 1.8274571805006588e-05,
+      "loss": 0.6803,
+      "step": 3750
+    },
+    {
+      "epoch": 0.5406366708162902,
+      "grad_norm": 11.762743949890137,
+      "learning_rate": 1.8221870882740448e-05,
+      "loss": 0.7343,
+      "step": 3800
+    },
+    {
+      "epoch": 0.5477503112217678,
+      "grad_norm": 3.884990692138672,
+      "learning_rate": 1.816916996047431e-05,
+      "loss": 0.7551,
+      "step": 3850
+    },
+    {
+      "epoch": 0.5548639516272452,
+      "grad_norm": 9.219622611999512,
+      "learning_rate": 1.8116469038208173e-05,
+      "loss": 0.764,
+      "step": 3900
+    },
+    {
+      "epoch": 0.5619775920327228,
+      "grad_norm": 7.808120250701904,
+      "learning_rate": 1.806376811594203e-05,
+      "loss": 0.745,
+      "step": 3950
+    },
+    {
+      "epoch": 0.5690912324382003,
+      "grad_norm": 7.188344955444336,
+      "learning_rate": 1.801106719367589e-05,
+      "loss": 0.7642,
+      "step": 4000
+    },
+    {
+      "epoch": 0.5762048728436777,
+      "grad_norm": 4.781955242156982,
+      "learning_rate": 1.795836627140975e-05,
+      "loss": 0.7089,
+      "step": 4050
+    },
+    {
+      "epoch": 0.5833185132491553,
+      "grad_norm": 4.489506721496582,
+      "learning_rate": 1.790566534914361e-05,
+      "loss": 0.7512,
+      "step": 4100
+    },
+    {
+      "epoch": 0.5904321536546328,
+      "grad_norm": 5.0979905128479,
+      "learning_rate": 1.785296442687747e-05,
+      "loss": 0.7753,
+      "step": 4150
+    },
+    {
+      "epoch": 0.5975457940601102,
+      "grad_norm": 5.302441596984863,
+      "learning_rate": 1.7800263504611332e-05,
+      "loss": 0.7412,
+      "step": 4200
+    },
+    {
+      "epoch": 0.6046594344655878,
+      "grad_norm": 4.942580223083496,
+      "learning_rate": 1.7747562582345192e-05,
+      "loss": 0.7474,
+      "step": 4250
+    },
+    {
+      "epoch": 0.6117730748710652,
+      "grad_norm": 4.685510158538818,
+      "learning_rate": 1.7694861660079053e-05,
+      "loss": 0.7173,
+      "step": 4300
+    },
+    {
+      "epoch": 0.6188867152765428,
+      "grad_norm": 4.8454999923706055,
+      "learning_rate": 1.7642160737812913e-05,
+      "loss": 0.7207,
+      "step": 4350
+    },
+    {
+      "epoch": 0.6260003556820203,
+      "grad_norm": 7.190736293792725,
+      "learning_rate": 1.7589459815546774e-05,
+      "loss": 0.754,
+      "step": 4400
+    },
+    {
+      "epoch": 0.6331139960874977,
+      "grad_norm": 7.0048675537109375,
+      "learning_rate": 1.7536758893280634e-05,
+      "loss": 0.7282,
+      "step": 4450
+    },
+    {
+      "epoch": 0.6402276364929753,
+      "grad_norm": 5.777256011962891,
+      "learning_rate": 1.7484057971014495e-05,
+      "loss": 0.6597,
+      "step": 4500
+    },
+    {
+      "epoch": 0.6473412768984528,
+      "grad_norm": 8.8043851852417,
+      "learning_rate": 1.7431357048748355e-05,
+      "loss": 0.6672,
+      "step": 4550
+    },
+    {
+      "epoch": 0.6544549173039302,
+      "grad_norm": 7.159558296203613,
+      "learning_rate": 1.7378656126482215e-05,
+      "loss": 0.7266,
+      "step": 4600
+    },
+    {
+      "epoch": 0.6615685577094078,
+      "grad_norm": 5.110541820526123,
+      "learning_rate": 1.7325955204216076e-05,
+      "loss": 0.6399,
+      "step": 4650
+    },
+    {
+      "epoch": 0.6686821981148853,
+      "grad_norm": 3.4727582931518555,
+      "learning_rate": 1.7273254281949936e-05,
+      "loss": 0.68,
+      "step": 4700
+    },
+    {
+      "epoch": 0.6757958385203628,
+      "grad_norm": 5.788763999938965,
+      "learning_rate": 1.7220553359683797e-05,
+      "loss": 0.7349,
+      "step": 4750
+    },
+    {
+      "epoch": 0.6829094789258403,
+      "grad_norm": 4.407217979431152,
+      "learning_rate": 1.7167852437417657e-05,
+      "loss": 0.695,
+      "step": 4800
+    },
+    {
+      "epoch": 0.6900231193313178,
+      "grad_norm": 4.685910224914551,
+      "learning_rate": 1.7115151515151514e-05,
+      "loss": 0.7816,
+      "step": 4850
+    },
+    {
+      "epoch": 0.6971367597367953,
+      "grad_norm": 2.9753873348236084,
+      "learning_rate": 1.7062450592885378e-05,
+      "loss": 0.6595,
+      "step": 4900
+    },
+    {
+      "epoch": 0.7042504001422728,
+      "grad_norm": 3.8326473236083984,
+      "learning_rate": 1.700974967061924e-05,
+      "loss": 0.6875,
+      "step": 4950
+    },
+    {
+      "epoch": 0.7113640405477503,
+      "grad_norm": 5.974637985229492,
+      "learning_rate": 1.69570487483531e-05,
+      "loss": 0.7414,
+      "step": 5000
+    },
+    {
+      "epoch": 0.7184776809532278,
+      "grad_norm": 3.4977550506591797,
+      "learning_rate": 1.6904347826086956e-05,
+      "loss": 0.6832,
+      "step": 5050
+    },
+    {
+      "epoch": 0.7255913213587053,
+      "grad_norm": 5.954402446746826,
+      "learning_rate": 1.6851646903820817e-05,
+      "loss": 0.7257,
+      "step": 5100
+    },
+    {
+      "epoch": 0.7327049617641829,
+      "grad_norm": 4.272510528564453,
+      "learning_rate": 1.679894598155468e-05,
+      "loss": 0.6808,
+      "step": 5150
+    },
+    {
+      "epoch": 0.7398186021696603,
+      "grad_norm": 4.155337810516357,
+      "learning_rate": 1.674624505928854e-05,
+      "loss": 0.7062,
+      "step": 5200
+    },
+    {
+      "epoch": 0.7469322425751378,
+      "grad_norm": 4.898423671722412,
+      "learning_rate": 1.6693544137022398e-05,
+      "loss": 0.6549,
+      "step": 5250
+    },
+    {
+      "epoch": 0.7540458829806154,
+      "grad_norm": 6.021943092346191,
+      "learning_rate": 1.664084321475626e-05,
+      "loss": 0.6886,
+      "step": 5300
+    },
+    {
+      "epoch": 0.7611595233860928,
+      "grad_norm": 4.547012805938721,
+      "learning_rate": 1.658814229249012e-05,
+      "loss": 0.6653,
+      "step": 5350
+    },
+    {
+      "epoch": 0.7682731637915703,
+      "grad_norm": 3.5163490772247314,
+      "learning_rate": 1.6535441370223983e-05,
+      "loss": 0.6549,
+      "step": 5400
+    },
+    {
+      "epoch": 0.7753868041970479,
+      "grad_norm": 4.255113124847412,
+      "learning_rate": 1.648274044795784e-05,
+      "loss": 0.6473,
+      "step": 5450
+    },
+    {
+      "epoch": 0.7825004446025253,
+      "grad_norm": 8.696562767028809,
+      "learning_rate": 1.64300395256917e-05,
+      "loss": 0.67,
+      "step": 5500
+    },
+    {
+      "epoch": 0.7896140850080029,
+      "grad_norm": 3.098172903060913,
+      "learning_rate": 1.637733860342556e-05,
+      "loss": 0.6626,
+      "step": 5550
+    },
+    {
+      "epoch": 0.7967277254134804,
+      "grad_norm": 5.950908660888672,
+      "learning_rate": 1.632463768115942e-05,
+      "loss": 0.7476,
+      "step": 5600
+    },
+    {
+      "epoch": 0.8038413658189578,
+      "grad_norm": 5.9870686531066895,
+      "learning_rate": 1.627193675889328e-05,
+      "loss": 0.7006,
+      "step": 5650
+    },
+    {
+      "epoch": 0.8109550062244354,
+      "grad_norm": 8.210447311401367,
+      "learning_rate": 1.6219235836627142e-05,
+      "loss": 0.6104,
+      "step": 5700
+    },
+    {
+      "epoch": 0.8180686466299129,
+      "grad_norm": 2.8877291679382324,
+      "learning_rate": 1.6166534914361002e-05,
+      "loss": 0.6183,
+      "step": 5750
+    },
+    {
+      "epoch": 0.8251822870353903,
+      "grad_norm": 3.5387609004974365,
+      "learning_rate": 1.6113833992094863e-05,
+      "loss": 0.6816,
+      "step": 5800
+    },
+    {
+      "epoch": 0.8322959274408679,
+      "grad_norm": 3.3335936069488525,
+      "learning_rate": 1.6061133069828723e-05,
+      "loss": 0.6948,
+      "step": 5850
+    },
+    {
+      "epoch": 0.8394095678463454,
+      "grad_norm": 3.515953302383423,
+      "learning_rate": 1.6008432147562584e-05,
+      "loss": 0.6664,
+      "step": 5900
+    },
+    {
+      "epoch": 0.8465232082518229,
+      "grad_norm": 2.2991325855255127,
+      "learning_rate": 1.5955731225296444e-05,
+      "loss": 0.6363,
+      "step": 5950
+    },
+    {
+      "epoch": 0.8536368486573004,
+      "grad_norm": 5.262946128845215,
+      "learning_rate": 1.5903030303030305e-05,
+      "loss": 0.7375,
+      "step": 6000
+    },
+    {
+      "epoch": 0.8607504890627778,
+      "grad_norm": 4.494121074676514,
+      "learning_rate": 1.5850329380764165e-05,
+      "loss": 0.6818,
+      "step": 6050
+    },
+    {
+      "epoch": 0.8678641294682554,
+      "grad_norm": 4.698576927185059,
+      "learning_rate": 1.5797628458498026e-05,
+      "loss": 0.7246,
+      "step": 6100
+    },
+    {
+      "epoch": 0.8749777698737329,
+      "grad_norm": 4.571809768676758,
+      "learning_rate": 1.5744927536231883e-05,
+      "loss": 0.6112,
+      "step": 6150
+    },
+    {
+      "epoch": 0.8820914102792103,
+      "grad_norm": 7.1781110763549805,
+      "learning_rate": 1.5692226613965746e-05,
+      "loss": 0.6593,
+      "step": 6200
+    },
+    {
+      "epoch": 0.8892050506846879,
+      "grad_norm": 5.867689609527588,
+      "learning_rate": 1.5639525691699607e-05,
+      "loss": 0.6491,
+      "step": 6250
+    },
+    {
+      "epoch": 0.8963186910901654,
+      "grad_norm": 5.427711009979248,
+      "learning_rate": 1.5586824769433467e-05,
+      "loss": 0.6419,
+      "step": 6300
+    },
+    {
+      "epoch": 0.903432331495643,
+      "grad_norm": 5.6072869300842285,
+      "learning_rate": 1.5534123847167328e-05,
+      "loss": 0.6443,
+      "step": 6350
+    },
+    {
+      "epoch": 0.9105459719011204,
+      "grad_norm": 4.355414867401123,
+      "learning_rate": 1.5481422924901185e-05,
+      "loss": 0.7021,
+      "step": 6400
+    },
+    {
+      "epoch": 0.9176596123065979,
+      "grad_norm": 4.039976596832275,
+      "learning_rate": 1.542872200263505e-05,
+      "loss": 0.7351,
+      "step": 6450
+    },
+    {
+      "epoch": 0.9247732527120754,
+      "grad_norm": 3.736238956451416,
+      "learning_rate": 1.537602108036891e-05,
+      "loss": 0.6629,
+      "step": 6500
+    },
+    {
+      "epoch": 0.9318868931175529,
+      "grad_norm": 3.91310453414917,
+      "learning_rate": 1.532332015810277e-05,
+      "loss": 0.6856,
+      "step": 6550
+    },
+    {
+      "epoch": 0.9390005335230304,
+      "grad_norm": 5.676443576812744,
+      "learning_rate": 1.5270619235836627e-05,
+      "loss": 0.6344,
+      "step": 6600
+    },
+    {
+      "epoch": 0.9461141739285079,
+      "grad_norm": 3.6127190589904785,
+      "learning_rate": 1.5217918313570487e-05,
+      "loss": 0.6845,
+      "step": 6650
+    },
+    {
+      "epoch": 0.9532278143339854,
+      "grad_norm": 4.957319259643555,
+      "learning_rate": 1.516521739130435e-05,
+      "loss": 0.7074,
+      "step": 6700
+    },
+    {
+      "epoch": 0.960341454739463,
+      "grad_norm": 3.351602554321289,
+      "learning_rate": 1.511251646903821e-05,
+      "loss": 0.6722,
+      "step": 6750
+    },
+    {
+      "epoch": 0.9674550951449404,
+      "grad_norm": 5.316050052642822,
+      "learning_rate": 1.505981554677207e-05,
+      "loss": 0.5993,
+      "step": 6800
+    },
+    {
+      "epoch": 0.9745687355504179,
+      "grad_norm": 3.5226705074310303,
+      "learning_rate": 1.5007114624505929e-05,
+      "loss": 0.6663,
+      "step": 6850
+    },
+    {
+      "epoch": 0.9816823759558955,
+      "grad_norm": 4.123106479644775,
+      "learning_rate": 1.495441370223979e-05,
+      "loss": 0.6865,
+      "step": 6900
+    },
+    {
+      "epoch": 0.9887960163613729,
+      "grad_norm": 3.942507743835449,
+      "learning_rate": 1.4901712779973652e-05,
+      "loss": 0.6535,
+      "step": 6950
+    },
+    {
+      "epoch": 0.9959096567668504,
+      "grad_norm": 4.765227317810059,
+      "learning_rate": 1.4849011857707512e-05,
+      "loss": 0.5895,
+      "step": 7000
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.6490142345428467,
+      "eval_runtime": 321.1223,
+      "eval_samples_per_second": 16.414,
+      "eval_steps_per_second": 2.052,
+      "step": 7029
+    },
+    {
+      "epoch": 1.0029877289703006,
+      "grad_norm": 3.913954496383667,
+      "learning_rate": 1.479631093544137e-05,
+      "loss": 0.5676,
+      "step": 7050
+    },
+    {
+      "epoch": 1.010101369375778,
+      "grad_norm": 2.70705246925354,
+      "learning_rate": 1.4743610013175231e-05,
+      "loss": 0.5268,
+      "step": 7100
+    },
+    {
+      "epoch": 1.0172150097812556,
+      "grad_norm": 5.550088882446289,
+      "learning_rate": 1.4690909090909092e-05,
+      "loss": 0.507,
+      "step": 7150
+    },
+    {
+      "epoch": 1.024328650186733,
+      "grad_norm": 6.876676082611084,
+      "learning_rate": 1.4638208168642954e-05,
+      "loss": 0.5285,
+      "step": 7200
+    },
+    {
+      "epoch": 1.0314422905922105,
+      "grad_norm": 2.911832809448242,
+      "learning_rate": 1.4585507246376813e-05,
+      "loss": 0.5357,
+      "step": 7250
+    },
+    {
+      "epoch": 1.0385559309976882,
+      "grad_norm": 5.395831108093262,
+      "learning_rate": 1.4532806324110673e-05,
+      "loss": 0.5442,
+      "step": 7300
+    },
+    {
+      "epoch": 1.0456695714031656,
+      "grad_norm": 8.814352035522461,
+      "learning_rate": 1.4480105401844533e-05,
+      "loss": 0.5394,
+      "step": 7350
+    },
+    {
+      "epoch": 1.052783211808643,
+      "grad_norm": 2.947053909301758,
+      "learning_rate": 1.4427404479578392e-05,
+      "loss": 0.5549,
+      "step": 7400
+    },
+    {
+      "epoch": 1.0598968522141206,
+      "grad_norm": 5.564446926116943,
+      "learning_rate": 1.4374703557312254e-05,
+      "loss": 0.5328,
+      "step": 7450
+    },
+    {
+      "epoch": 1.067010492619598,
+      "grad_norm": 4.452253341674805,
+      "learning_rate": 1.4322002635046115e-05,
+      "loss": 0.5197,
+      "step": 7500
+    },
+    {
+      "epoch": 1.0741241330250757,
+      "grad_norm": 3.506099224090576,
+      "learning_rate": 1.4269301712779975e-05,
+      "loss": 0.5509,
+      "step": 7550
+    },
+    {
+      "epoch": 1.0812377734305532,
+      "grad_norm": 4.477612018585205,
+      "learning_rate": 1.4216600790513834e-05,
+      "loss": 0.5005,
+      "step": 7600
+    },
+    {
+      "epoch": 1.0883514138360306,
+      "grad_norm": 4.912115097045898,
+      "learning_rate": 1.4163899868247694e-05,
+      "loss": 0.5351,
+      "step": 7650
+    },
+    {
+      "epoch": 1.095465054241508,
+      "grad_norm": 4.499685764312744,
+      "learning_rate": 1.4111198945981557e-05,
+      "loss": 0.5218,
+      "step": 7700
+    },
+    {
+      "epoch": 1.1025786946469855,
+      "grad_norm": 4.6450934410095215,
+      "learning_rate": 1.4058498023715417e-05,
+      "loss": 0.5195,
+      "step": 7750
+    },
+    {
+      "epoch": 1.109692335052463,
+      "grad_norm": 4.883353233337402,
+      "learning_rate": 1.4005797101449276e-05,
+      "loss": 0.5464,
+      "step": 7800
+    },
+    {
+      "epoch": 1.1168059754579407,
+      "grad_norm": 3.312135934829712,
+      "learning_rate": 1.3953096179183136e-05,
+      "loss": 0.4766,
+      "step": 7850
+    },
+    {
+      "epoch": 1.1239196158634182,
+      "grad_norm": 2.180053234100342,
+      "learning_rate": 1.3900395256916997e-05,
+      "loss": 0.5742,
+      "step": 7900
+    },
+    {
+      "epoch": 1.1310332562688956,
+      "grad_norm": 4.708530426025391,
+      "learning_rate": 1.3847694334650859e-05,
+      "loss": 0.5499,
+      "step": 7950
+    },
+    {
+      "epoch": 1.138146896674373,
+      "grad_norm": 3.369694709777832,
+      "learning_rate": 1.379499341238472e-05,
+      "loss": 0.4839,
+      "step": 8000
+    },
+    {
+      "epoch": 1.1452605370798505,
+      "grad_norm": 3.1507697105407715,
+      "learning_rate": 1.3742292490118578e-05,
+      "loss": 0.5118,
+      "step": 8050
+    },
+    {
+      "epoch": 1.1523741774853282,
+      "grad_norm": 3.773084878921509,
+      "learning_rate": 1.3689591567852438e-05,
+      "loss": 0.5406,
+      "step": 8100
+    },
+    {
+      "epoch": 1.1594878178908057,
+      "grad_norm": 3.446446180343628,
+      "learning_rate": 1.3636890645586299e-05,
+      "loss": 0.5175,
+      "step": 8150
+    },
+    {
+      "epoch": 1.1666014582962831,
+      "grad_norm": 5.0827155113220215,
+      "learning_rate": 1.3584189723320161e-05,
+      "loss": 0.4963,
+      "step": 8200
+    },
+    {
+      "epoch": 1.1737150987017606,
+      "grad_norm": 5.854413032531738,
+      "learning_rate": 1.353148880105402e-05,
+      "loss": 0.6038,
+      "step": 8250
+    },
+    {
+      "epoch": 1.180828739107238,
+      "grad_norm": 4.135270595550537,
+      "learning_rate": 1.347878787878788e-05,
+      "loss": 0.5876,
+      "step": 8300
+    },
+    {
+      "epoch": 1.1879423795127155,
+      "grad_norm": 4.9584197998046875,
+      "learning_rate": 1.342608695652174e-05,
+      "loss": 0.5272,
+      "step": 8350
+    },
+    {
+      "epoch": 1.1950560199181932,
+      "grad_norm": 2.152341842651367,
+      "learning_rate": 1.33733860342556e-05,
+      "loss": 0.5146,
+      "step": 8400
+    },
+    {
+      "epoch": 1.2021696603236707,
+      "grad_norm": 3.346052885055542,
+      "learning_rate": 1.3320685111989462e-05,
+      "loss": 0.4661,
+      "step": 8450
+    },
+    {
+      "epoch": 1.2092833007291481,
+      "grad_norm": 4.768698692321777,
+      "learning_rate": 1.3267984189723322e-05,
+      "loss": 0.5354,
+      "step": 8500
+    },
+    {
+      "epoch": 1.2163969411346256,
+      "grad_norm": 5.182507514953613,
+      "learning_rate": 1.3215283267457183e-05,
+      "loss": 0.4845,
+      "step": 8550
+    },
+    {
+      "epoch": 1.223510581540103,
+      "grad_norm": 4.584784030914307,
+      "learning_rate": 1.3162582345191041e-05,
+      "loss": 0.5261,
+      "step": 8600
+    },
+    {
+      "epoch": 1.2306242219455807,
+      "grad_norm": 3.520695209503174,
+      "learning_rate": 1.3110935441370225e-05,
+      "loss": 0.5735,
+      "step": 8650
+    },
+    {
+      "epoch": 1.2377378623510582,
+      "grad_norm": 5.633211612701416,
+      "learning_rate": 1.3058234519104085e-05,
+      "loss": 0.516,
+      "step": 8700
+    },
+    {
+      "epoch": 1.2448515027565357,
+      "grad_norm": 5.559876441955566,
+      "learning_rate": 1.3005533596837944e-05,
+      "loss": 0.5597,
+      "step": 8750
+    },
+    {
+      "epoch": 1.2519651431620131,
+      "grad_norm": 4.474053382873535,
+      "learning_rate": 1.2952832674571806e-05,
+      "loss": 0.5171,
+      "step": 8800
+    },
+    {
+      "epoch": 1.2590787835674906,
+      "grad_norm": 3.866060733795166,
+      "learning_rate": 1.2900131752305667e-05,
+      "loss": 0.5057,
+      "step": 8850
+    },
+    {
+      "epoch": 1.266192423972968,
+      "grad_norm": 5.632756233215332,
+      "learning_rate": 1.2847430830039527e-05,
+      "loss": 0.5514,
+      "step": 8900
+    },
+    {
+      "epoch": 1.2733060643784457,
+      "grad_norm": 4.323223114013672,
+      "learning_rate": 1.2794729907773386e-05,
+      "loss": 0.5052,
+      "step": 8950
+    },
+    {
+      "epoch": 1.2804197047839232,
+      "grad_norm": 3.6960432529449463,
+      "learning_rate": 1.2742028985507246e-05,
+      "loss": 0.5295,
+      "step": 9000
+    },
+    {
+      "epoch": 1.2875333451894007,
+      "grad_norm": 3.777792453765869,
+      "learning_rate": 1.2689328063241108e-05,
+      "loss": 0.49,
+      "step": 9050
+    },
+    {
+      "epoch": 1.2946469855948781,
+      "grad_norm": 2.103578805923462,
+      "learning_rate": 1.2636627140974969e-05,
+      "loss": 0.5437,
+      "step": 9100
+    },
+    {
+      "epoch": 1.3017606260003558,
+      "grad_norm": 5.730409622192383,
+      "learning_rate": 1.258392621870883e-05,
+      "loss": 0.532,
+      "step": 9150
+    },
+    {
+      "epoch": 1.3088742664058333,
+      "grad_norm": 6.736247539520264,
+      "learning_rate": 1.2532279314888011e-05,
+      "loss": 0.554,
+      "step": 9200
+    },
+    {
+      "epoch": 1.3159879068113107,
+      "grad_norm": 2.0751116275787354,
+      "learning_rate": 1.2479578392621872e-05,
+      "loss": 0.5128,
+      "step": 9250
+    },
+    {
+      "epoch": 1.3231015472167882,
+      "grad_norm": 2.6663055419921875,
+      "learning_rate": 1.2426877470355732e-05,
+      "loss": 0.5251,
+      "step": 9300
+    },
+    {
+      "epoch": 1.3302151876222656,
+      "grad_norm": 4.778317928314209,
+      "learning_rate": 1.2374176548089594e-05,
+      "loss": 0.5219,
+      "step": 9350
+    },
+    {
+      "epoch": 1.337328828027743,
+      "grad_norm": 5.950105667114258,
+      "learning_rate": 1.2321475625823453e-05,
+      "loss": 0.5192,
+      "step": 9400
+    },
+    {
+      "epoch": 1.3444424684332206,
+      "grad_norm": 3.5832479000091553,
+      "learning_rate": 1.2268774703557313e-05,
+      "loss": 0.5185,
+      "step": 9450
+    },
+    {
+      "epoch": 1.3515561088386983,
+      "grad_norm": 2.7390801906585693,
+      "learning_rate": 1.2216073781291174e-05,
+      "loss": 0.5375,
+      "step": 9500
+    },
+    {
+      "epoch": 1.3586697492441757,
+      "grad_norm": 2.688469648361206,
+      "learning_rate": 1.2163372859025033e-05,
+      "loss": 0.5352,
+      "step": 9550
+    },
+    {
+      "epoch": 1.3657833896496532,
+      "grad_norm": 4.956066608428955,
+      "learning_rate": 1.2110671936758893e-05,
+      "loss": 0.5341,
+      "step": 9600
+    },
+    {
+      "epoch": 1.3728970300551306,
+      "grad_norm": 5.483800411224365,
+      "learning_rate": 1.2057971014492755e-05,
+      "loss": 0.5644,
+      "step": 9650
+    },
+    {
+      "epoch": 1.3800106704606083,
+      "grad_norm": 4.272674083709717,
+      "learning_rate": 1.2005270092226616e-05,
+      "loss": 0.5044,
+      "step": 9700
+    },
+    {
+      "epoch": 1.3871243108660858,
+      "grad_norm": 2.5622191429138184,
+      "learning_rate": 1.1952569169960474e-05,
+      "loss": 0.5171,
+      "step": 9750
+    },
+    {
+      "epoch": 1.3942379512715632,
+      "grad_norm": 7.005847454071045,
+      "learning_rate": 1.1899868247694335e-05,
+      "loss": 0.5156,
+      "step": 9800
+    },
+    {
+      "epoch": 1.4013515916770407,
+      "grad_norm": 4.284529685974121,
+      "learning_rate": 1.1847167325428195e-05,
+      "loss": 0.5833,
+      "step": 9850
+    },
+    {
+      "epoch": 1.4084652320825182,
+      "grad_norm": 3.930690288543701,
+      "learning_rate": 1.1794466403162057e-05,
+      "loss": 0.5219,
+      "step": 9900
+    },
+    {
+      "epoch": 1.4155788724879956,
+      "grad_norm": 4.899880886077881,
+      "learning_rate": 1.1741765480895916e-05,
+      "loss": 0.5605,
+      "step": 9950
+    },
+    {
+      "epoch": 1.4226925128934733,
+      "grad_norm": 5.612455368041992,
+      "learning_rate": 1.1689064558629777e-05,
+      "loss": 0.5408,
+      "step": 10000
+    },
+    {
+      "epoch": 1.4298061532989508,
+      "grad_norm": 2.502410411834717,
+      "learning_rate": 1.1636363636363637e-05,
+      "loss": 0.5052,
+      "step": 10050
+    },
+    {
+      "epoch": 1.4369197937044282,
+      "grad_norm": 4.965051174163818,
+      "learning_rate": 1.1583662714097496e-05,
+      "loss": 0.5036,
+      "step": 10100
+    },
+    {
+      "epoch": 1.4440334341099057,
+      "grad_norm": 4.5783371925354,
+      "learning_rate": 1.1530961791831358e-05,
+      "loss": 0.5474,
+      "step": 10150
+    },
+    {
+      "epoch": 1.4511470745153834,
+      "grad_norm": 5.443408012390137,
+      "learning_rate": 1.1478260869565218e-05,
+      "loss": 0.5018,
+      "step": 10200
+    },
+    {
+      "epoch": 1.4582607149208608,
+      "grad_norm": 5.504450798034668,
+      "learning_rate": 1.1425559947299079e-05,
+      "loss": 0.4877,
+      "step": 10250
+    },
+    {
+      "epoch": 1.4653743553263383,
+      "grad_norm": 3.819101572036743,
+      "learning_rate": 1.137285902503294e-05,
+      "loss": 0.526,
+      "step": 10300
+    },
+    {
+      "epoch": 1.4724879957318158,
+      "grad_norm": 5.407709121704102,
+      "learning_rate": 1.1320158102766798e-05,
+      "loss": 0.4949,
+      "step": 10350
+    },
+    {
+      "epoch": 1.4796016361372932,
+      "grad_norm": 4.124641418457031,
+      "learning_rate": 1.126745718050066e-05,
+      "loss": 0.5923,
+      "step": 10400
+    },
+    {
+      "epoch": 1.4867152765427707,
+      "grad_norm": 4.266039848327637,
+      "learning_rate": 1.121475625823452e-05,
+      "loss": 0.5545,
+      "step": 10450
+    },
+    {
+      "epoch": 1.4938289169482482,
+      "grad_norm": 4.402031421661377,
+      "learning_rate": 1.1162055335968381e-05,
+      "loss": 0.505,
+      "step": 10500
+    },
+    {
+      "epoch": 1.5009425573537256,
+      "grad_norm": 6.378383159637451,
+      "learning_rate": 1.110935441370224e-05,
+      "loss": 0.5015,
+      "step": 10550
+    },
+    {
+      "epoch": 1.5080561977592033,
+      "grad_norm": 4.023849964141846,
+      "learning_rate": 1.10566534914361e-05,
+      "loss": 0.5371,
+      "step": 10600
+    },
+    {
+      "epoch": 1.5151698381646808,
+      "grad_norm": 4.202026844024658,
+      "learning_rate": 1.1003952569169962e-05,
+      "loss": 0.5527,
+      "step": 10650
+    },
+    {
+      "epoch": 1.5222834785701584,
+      "grad_norm": 4.9058685302734375,
+      "learning_rate": 1.0951251646903823e-05,
+      "loss": 0.5193,
+      "step": 10700
+    },
+    {
+      "epoch": 1.529397118975636,
+      "grad_norm": 5.402507781982422,
+      "learning_rate": 1.0898550724637682e-05,
+      "loss": 0.542,
+      "step": 10750
+    },
+    {
+      "epoch": 1.5365107593811134,
+      "grad_norm": 4.229278087615967,
+      "learning_rate": 1.0845849802371542e-05,
+      "loss": 0.5284,
+      "step": 10800
+    },
+    {
+      "epoch": 1.5436243997865908,
+      "grad_norm": 3.6707725524902344,
+      "learning_rate": 1.0793148880105403e-05,
+      "loss": 0.561,
+      "step": 10850
+    },
+    {
+      "epoch": 1.5507380401920683,
+      "grad_norm": 3.6205899715423584,
+      "learning_rate": 1.0740447957839265e-05,
+      "loss": 0.4981,
+      "step": 10900
+    },
+    {
+      "epoch": 1.5578516805975458,
+      "grad_norm": 4.793003559112549,
+      "learning_rate": 1.0687747035573123e-05,
+      "loss": 0.5062,
+      "step": 10950
+    },
+    {
+      "epoch": 1.5649653210030232,
+      "grad_norm": 3.2560999393463135,
+      "learning_rate": 1.0635046113306984e-05,
+      "loss": 0.5305,
+      "step": 11000
+    },
+    {
+      "epoch": 1.5720789614085007,
+      "grad_norm": 4.039142608642578,
+      "learning_rate": 1.0582345191040844e-05,
+      "loss": 0.5412,
+      "step": 11050
+    },
+    {
+      "epoch": 1.5791926018139784,
+      "grad_norm": 4.209068298339844,
+      "learning_rate": 1.0529644268774703e-05,
+      "loss": 0.5014,
+      "step": 11100
+    },
+    {
+      "epoch": 1.5863062422194558,
+      "grad_norm": 3.390684127807617,
+      "learning_rate": 1.0476943346508565e-05,
+      "loss": 0.5605,
+      "step": 11150
+    },
+    {
+      "epoch": 1.5934198826249333,
+      "grad_norm": 3.7350914478302,
+      "learning_rate": 1.0424242424242426e-05,
+      "loss": 0.5021,
+      "step": 11200
+    },
+    {
+      "epoch": 1.600533523030411,
+      "grad_norm": 3.5504932403564453,
+      "learning_rate": 1.0371541501976286e-05,
+      "loss": 0.4967,
+      "step": 11250
+    },
+    {
+      "epoch": 1.6076471634358884,
+      "grad_norm": 3.8655166625976562,
+      "learning_rate": 1.0318840579710145e-05,
+      "loss": 0.4857,
+      "step": 11300
+    },
+    {
+      "epoch": 1.6147608038413659,
+      "grad_norm": 6.647157669067383,
+      "learning_rate": 1.0266139657444005e-05,
+      "loss": 0.5033,
+      "step": 11350
+    },
+    {
+      "epoch": 1.6218744442468433,
+      "grad_norm": 5.384164810180664,
+      "learning_rate": 1.0213438735177868e-05,
+      "loss": 0.5486,
+      "step": 11400
+    },
+    {
+      "epoch": 1.6289880846523208,
+      "grad_norm": 4.699065685272217,
+      "learning_rate": 1.016179183135705e-05,
+      "loss": 0.5565,
+      "step": 11450
+    },
+    {
+      "epoch": 1.6361017250577983,
+      "grad_norm": 4.786224365234375,
+      "learning_rate": 1.010909090909091e-05,
+      "loss": 0.5155,
+      "step": 11500
+    },
+    {
+      "epoch": 1.6432153654632757,
+      "grad_norm": 4.30881929397583,
+      "learning_rate": 1.005638998682477e-05,
+      "loss": 0.5303,
+      "step": 11550
+    },
+    {
+      "epoch": 1.6503290058687532,
+      "grad_norm": 4.938950538635254,
+      "learning_rate": 1.000368906455863e-05,
+      "loss": 0.4891,
+      "step": 11600
+    },
+    {
+      "epoch": 1.6574426462742309,
+      "grad_norm": 4.875250339508057,
+      "learning_rate": 9.950988142292491e-06,
+      "loss": 0.531,
+      "step": 11650
+    },
+    {
+      "epoch": 1.6645562866797083,
+      "grad_norm": 5.96549654006958,
+      "learning_rate": 9.898287220026352e-06,
+      "loss": 0.5092,
+      "step": 11700
+    },
+    {
+      "epoch": 1.6716699270851858,
+      "grad_norm": 5.401658535003662,
+      "learning_rate": 9.845586297760212e-06,
+      "loss": 0.5018,
+      "step": 11750
+    },
+    {
+      "epoch": 1.6787835674906635,
+      "grad_norm": 4.49137020111084,
+      "learning_rate": 9.792885375494072e-06,
+      "loss": 0.4877,
+      "step": 11800
+    },
+    {
+      "epoch": 1.685897207896141,
+      "grad_norm": 3.0070083141326904,
+      "learning_rate": 9.740184453227933e-06,
+      "loss": 0.4884,
+      "step": 11850
+    },
+    {
+      "epoch": 1.6930108483016184,
+      "grad_norm": 3.979234218597412,
+      "learning_rate": 9.687483530961793e-06,
+      "loss": 0.5013,
+      "step": 11900
+    },
+    {
+      "epoch": 1.7001244887070959,
+      "grad_norm": 3.876878499984741,
+      "learning_rate": 9.634782608695654e-06,
+      "loss": 0.5322,
+      "step": 11950
+    },
+    {
+      "epoch": 1.7072381291125733,
+      "grad_norm": 5.199528694152832,
+      "learning_rate": 9.582081686429513e-06,
+      "loss": 0.5052,
+      "step": 12000
+    },
+    {
+      "epoch": 1.7143517695180508,
+      "grad_norm": 3.97282075881958,
+      "learning_rate": 9.529380764163375e-06,
+      "loss": 0.4887,
+      "step": 12050
+    },
+    {
+      "epoch": 1.7214654099235283,
+      "grad_norm": 4.776126384735107,
+      "learning_rate": 9.476679841897233e-06,
+      "loss": 0.5384,
+      "step": 12100
+    },
+    {
+      "epoch": 1.7285790503290057,
+      "grad_norm": 6.539297580718994,
+      "learning_rate": 9.423978919631094e-06,
+      "loss": 0.5283,
+      "step": 12150
+    },
+    {
+      "epoch": 1.7356926907344834,
+      "grad_norm": 5.380597114562988,
+      "learning_rate": 9.371277997364954e-06,
+      "loss": 0.5481,
+      "step": 12200
+    },
+    {
+      "epoch": 1.7428063311399609,
+      "grad_norm": 4.773643493652344,
+      "learning_rate": 9.318577075098815e-06,
+      "loss": 0.4815,
+      "step": 12250
+    },
+    {
+      "epoch": 1.7499199715454383,
+      "grad_norm": 1.9399245977401733,
+      "learning_rate": 9.265876152832675e-06,
+      "loss": 0.515,
+      "step": 12300
+    },
+    {
+      "epoch": 1.757033611950916,
+      "grad_norm": 3.586733102798462,
+      "learning_rate": 9.213175230566536e-06,
+      "loss": 0.4856,
+      "step": 12350
+    },
+    {
+      "epoch": 1.7641472523563935,
+      "grad_norm": 3.9114813804626465,
+      "learning_rate": 9.160474308300396e-06,
+      "loss": 0.5114,
+      "step": 12400
+    },
+    {
+      "epoch": 1.771260892761871,
+      "grad_norm": 3.8798410892486572,
+      "learning_rate": 9.107773386034257e-06,
+      "loss": 0.4939,
+      "step": 12450
+    },
+    {
+      "epoch": 1.7783745331673484,
+      "grad_norm": 2.8337159156799316,
+      "learning_rate": 9.055072463768117e-06,
+      "loss": 0.4807,
+      "step": 12500
+    },
+    {
+      "epoch": 1.7854881735728259,
+      "grad_norm": 4.102101802825928,
+      "learning_rate": 9.002371541501978e-06,
+      "loss": 0.5632,
+      "step": 12550
+    },
+    {
+      "epoch": 1.7926018139783033,
+      "grad_norm": 2.918848991394043,
+      "learning_rate": 8.949670619235838e-06,
+      "loss": 0.5012,
+      "step": 12600
+    },
+    {
+      "epoch": 1.7997154543837808,
+      "grad_norm": 3.708796739578247,
+      "learning_rate": 8.896969696969697e-06,
+      "loss": 0.5159,
+      "step": 12650
+    },
+    {
+      "epoch": 1.8068290947892582,
+      "grad_norm": 4.967949867248535,
+      "learning_rate": 8.844268774703559e-06,
+      "loss": 0.5037,
+      "step": 12700
+    },
+    {
+      "epoch": 1.813942735194736,
+      "grad_norm": 2.7646732330322266,
+      "learning_rate": 8.791567852437418e-06,
+      "loss": 0.4842,
+      "step": 12750
+    },
+    {
+      "epoch": 1.8210563756002134,
+      "grad_norm": 3.667774200439453,
+      "learning_rate": 8.73886693017128e-06,
+      "loss": 0.4744,
+      "step": 12800
+    },
+    {
+      "epoch": 1.828170016005691,
+      "grad_norm": 5.304251670837402,
+      "learning_rate": 8.686166007905139e-06,
+      "loss": 0.521,
+      "step": 12850
+    },
+    {
+      "epoch": 1.8352836564111685,
+      "grad_norm": 4.663897514343262,
+      "learning_rate": 8.633465085638999e-06,
+      "loss": 0.5143,
+      "step": 12900
+    },
+    {
+      "epoch": 1.842397296816646,
+      "grad_norm": 6.192160129547119,
+      "learning_rate": 8.58076416337286e-06,
+      "loss": 0.5489,
+      "step": 12950
+    },
+    {
+      "epoch": 1.8495109372221235,
+      "grad_norm": 3.5136263370513916,
+      "learning_rate": 8.52806324110672e-06,
+      "loss": 0.5124,
+      "step": 13000
+    },
+    {
+      "epoch": 1.856624577627601,
+      "grad_norm": 3.0261971950531006,
+      "learning_rate": 8.47536231884058e-06,
+      "loss": 0.5295,
+      "step": 13050
+    },
+    {
+      "epoch": 1.8637382180330784,
+      "grad_norm": 3.9308557510375977,
+      "learning_rate": 8.42266139657444e-06,
+      "loss": 0.5209,
+      "step": 13100
+    },
+    {
+      "epoch": 1.8708518584385558,
+      "grad_norm": 4.964597225189209,
+      "learning_rate": 8.369960474308301e-06,
+      "loss": 0.5187,
+      "step": 13150
+    },
+    {
+      "epoch": 1.8779654988440333,
+      "grad_norm": 5.161786079406738,
+      "learning_rate": 8.317259552042162e-06,
+      "loss": 0.5207,
+      "step": 13200
+    },
+    {
+      "epoch": 1.885079139249511,
+      "grad_norm": 2.5828144550323486,
+      "learning_rate": 8.264558629776022e-06,
+      "loss": 0.4598,
+      "step": 13250
+    },
+    {
+      "epoch": 1.8921927796549884,
+      "grad_norm": 3.169773817062378,
+      "learning_rate": 8.211857707509883e-06,
+      "loss": 0.531,
+      "step": 13300
+    },
+    {
+      "epoch": 1.899306420060466,
+      "grad_norm": 3.6591272354125977,
+      "learning_rate": 8.159156785243743e-06,
+      "loss": 0.5226,
+      "step": 13350
+    },
+    {
+      "epoch": 1.9064200604659436,
+      "grad_norm": 2.735844850540161,
+      "learning_rate": 8.106455862977602e-06,
+      "loss": 0.5808,
+      "step": 13400
+    },
+    {
+      "epoch": 1.913533700871421,
+      "grad_norm": 5.649113655090332,
+      "learning_rate": 8.053754940711464e-06,
+      "loss": 0.5154,
+      "step": 13450
+    },
+    {
+      "epoch": 1.9206473412768985,
+      "grad_norm": 2.830268621444702,
+      "learning_rate": 8.001054018445323e-06,
+      "loss": 0.5017,
+      "step": 13500
+    },
+    {
+      "epoch": 1.927760981682376,
+      "grad_norm": 5.496303558349609,
+      "learning_rate": 7.948353096179183e-06,
+      "loss": 0.4824,
+      "step": 13550
+    },
+    {
+      "epoch": 1.9348746220878534,
+      "grad_norm": 3.1559128761291504,
+      "learning_rate": 7.895652173913044e-06,
+      "loss": 0.5465,
+      "step": 13600
+    },
+    {
+      "epoch": 1.941988262493331,
+      "grad_norm": 3.265587091445923,
+      "learning_rate": 7.842951251646904e-06,
+      "loss": 0.4673,
+      "step": 13650
+    },
+    {
+      "epoch": 1.9491019028988084,
+      "grad_norm": 6.156945705413818,
+      "learning_rate": 7.790250329380764e-06,
+      "loss": 0.5558,
+      "step": 13700
+    },
+    {
+      "epoch": 1.9562155433042858,
+      "grad_norm": 7.1588029861450195,
+      "learning_rate": 7.737549407114625e-06,
+      "loss": 0.5013,
+      "step": 13750
+    },
+    {
+      "epoch": 1.9633291837097635,
+      "grad_norm": 3.9235057830810547,
+      "learning_rate": 7.684848484848485e-06,
+      "loss": 0.5214,
+      "step": 13800
+    },
+    {
+      "epoch": 1.970442824115241,
+      "grad_norm": 3.9342610836029053,
+      "learning_rate": 7.632147562582346e-06,
+      "loss": 0.5147,
+      "step": 13850
+    },
+    {
+      "epoch": 1.9775564645207184,
+      "grad_norm": 2.9213314056396484,
+      "learning_rate": 7.579446640316206e-06,
+      "loss": 0.4756,
+      "step": 13900
+    },
+    {
+      "epoch": 1.984670104926196,
+      "grad_norm": 2.211418390274048,
+      "learning_rate": 7.526745718050067e-06,
+      "loss": 0.4862,
+      "step": 13950
+    },
+    {
+      "epoch": 1.9917837453316736,
+      "grad_norm": 3.4956717491149902,
+      "learning_rate": 7.474044795783927e-06,
+      "loss": 0.5235,
+      "step": 14000
+    },
+    {
+      "epoch": 1.998897385737151,
+      "grad_norm": 4.755037784576416,
+      "learning_rate": 7.421343873517787e-06,
+      "loss": 0.5176,
+      "step": 14050
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6273267865180969,
+      "eval_runtime": 321.2086,
+      "eval_samples_per_second": 16.41,
+      "eval_steps_per_second": 2.052,
+      "step": 14058
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 21084,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 7.608571407630336e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b38f53c888a9c457e9e1a092c06ddd45fb62219eaec68a8e171773f29b39d4db
+size 5713