Update scenario_cot_ratio_1.5B

.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
+scenario_cot_ratio_1.5B/tokenizer.json filter=lfs diff=lfs merge=lfs -text

scenario_cot_ratio_1.5B/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

scenario_cot_ratio_1.5B/all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 2.9960134927936215,
+    "eval_loss": 0.7925666570663452,
+    "eval_runtime": 3.2324,
+    "eval_samples_per_second": 24.44,
+    "eval_steps_per_second": 6.187,
+    "total_flos": 337549181616128.0,
+    "train_loss": 0.809739662817909,
+    "train_runtime": 19632.9853,
+    "train_samples_per_second": 5.978,
+    "train_steps_per_second": 0.062
+}

scenario_cot_ratio_1.5B/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "/mnt/public/code/zhaojihao/models/Qwen2.5-1.5B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 8960,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "qwen2",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.1",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

scenario_cot_ratio_1.5B/eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.9960134927936215,
+    "eval_loss": 0.7925666570663452,
+    "eval_runtime": 3.2324,
+    "eval_samples_per_second": 24.44,
+    "eval_steps_per_second": 6.187
+}

scenario_cot_ratio_1.5B/generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.46.1"
+}

scenario_cot_ratio_1.5B/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fe8e186b40b3795ff779fb1fa265143d8dead77ce65bc0e92f1424f9ea9168e
+size 3087467144

scenario_cot_ratio_1.5B/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
+  }
+}

scenario_cot_ratio_1.5B/tokenizer.json

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

scenario_cot_ratio_1.5B/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\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    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) 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{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

scenario_cot_ratio_1.5B/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9960134927936215,
+    "total_flos": 337549181616128.0,
+    "train_loss": 0.809739662817909,
+    "train_runtime": 19632.9853,
+    "train_samples_per_second": 5.978,
+    "train_steps_per_second": 0.062
+}

scenario_cot_ratio_1.5B/trainer_state.json

@@ -0,0 +1,1384 @@
+{
+  "best_metric": 0.7925666570663452,
+  "best_model_checkpoint": "/mnt/public/code/zhaojihao/train_model/scenario_train_40K_cot_ratio_qwen25_15B_3/checkpoint-1200",
+  "epoch": 2.9960134927936215,
+  "eval_steps": 20,
+  "global_step": 1221,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.024532352039251762,
+      "grad_norm": 3.9774076796797275,
+      "learning_rate": 8.130081300813009e-07,
+      "loss": 1.5356,
+      "step": 10
+    },
+    {
+      "epoch": 0.049064704078503524,
+      "grad_norm": 2.733650252044082,
+      "learning_rate": 1.6260162601626018e-06,
+      "loss": 1.494,
+      "step": 20
+    },
+    {
+      "epoch": 0.049064704078503524,
+      "eval_loss": 1.4565786123275757,
+      "eval_runtime": 3.3847,
+      "eval_samples_per_second": 23.34,
+      "eval_steps_per_second": 5.909,
+      "step": 20
+    },
+    {
+      "epoch": 0.07359705611775529,
+      "grad_norm": 1.9141519197953465,
+      "learning_rate": 2.4390243902439027e-06,
+      "loss": 1.3603,
+      "step": 30
+    },
+    {
+      "epoch": 0.09812940815700705,
+      "grad_norm": 1.0928465548028006,
+      "learning_rate": 3.2520325203252037e-06,
+      "loss": 1.2303,
+      "step": 40
+    },
+    {
+      "epoch": 0.09812940815700705,
+      "eval_loss": 1.2004287242889404,
+      "eval_runtime": 3.245,
+      "eval_samples_per_second": 24.345,
+      "eval_steps_per_second": 6.163,
+      "step": 40
+    },
+    {
+      "epoch": 0.12266176019625881,
+      "grad_norm": 0.8318905756386776,
+      "learning_rate": 4.0650406504065046e-06,
+      "loss": 1.1353,
+      "step": 50
+    },
+    {
+      "epoch": 0.14719411223551057,
+      "grad_norm": 0.7096286697472151,
+      "learning_rate": 4.8780487804878055e-06,
+      "loss": 1.0715,
+      "step": 60
+    },
+    {
+      "epoch": 0.14719411223551057,
+      "eval_loss": 1.0666805505752563,
+      "eval_runtime": 3.2597,
+      "eval_samples_per_second": 24.236,
+      "eval_steps_per_second": 6.136,
+      "step": 60
+    },
+    {
+      "epoch": 0.17172646427476235,
+      "grad_norm": 0.6461933135743275,
+      "learning_rate": 5.691056910569106e-06,
+      "loss": 1.0212,
+      "step": 70
+    },
+    {
+      "epoch": 0.1962588163140141,
+      "grad_norm": 0.6450560615208996,
+      "learning_rate": 6.504065040650407e-06,
+      "loss": 0.996,
+      "step": 80
+    },
+    {
+      "epoch": 0.1962588163140141,
+      "eval_loss": 1.0017997026443481,
+      "eval_runtime": 3.2596,
+      "eval_samples_per_second": 24.236,
+      "eval_steps_per_second": 6.136,
+      "step": 80
+    },
+    {
+      "epoch": 0.22079116835326587,
+      "grad_norm": 0.6339481212206293,
+      "learning_rate": 7.317073170731707e-06,
+      "loss": 0.9698,
+      "step": 90
+    },
+    {
+      "epoch": 0.24532352039251762,
+      "grad_norm": 0.6466194203559675,
+      "learning_rate": 8.130081300813009e-06,
+      "loss": 0.958,
+      "step": 100
+    },
+    {
+      "epoch": 0.24532352039251762,
+      "eval_loss": 0.9641426801681519,
+      "eval_runtime": 3.2724,
+      "eval_samples_per_second": 24.141,
+      "eval_steps_per_second": 6.112,
+      "step": 100
+    },
+    {
+      "epoch": 0.2698558724317694,
+      "grad_norm": 0.6764514370839726,
+      "learning_rate": 8.94308943089431e-06,
+      "loss": 0.9381,
+      "step": 110
+    },
+    {
+      "epoch": 0.29438822447102114,
+      "grad_norm": 0.7354109361398842,
+      "learning_rate": 9.756097560975611e-06,
+      "loss": 0.9286,
+      "step": 120
+    },
+    {
+      "epoch": 0.29438822447102114,
+      "eval_loss": 0.938707172870636,
+      "eval_runtime": 3.2183,
+      "eval_samples_per_second": 24.547,
+      "eval_steps_per_second": 6.215,
+      "step": 120
+    },
+    {
+      "epoch": 0.31892057651027295,
+      "grad_norm": 0.6795152732594263,
+      "learning_rate": 9.998997194663937e-06,
+      "loss": 0.9157,
+      "step": 130
+    },
+    {
+      "epoch": 0.3434529285495247,
+      "grad_norm": 0.6748829559062427,
+      "learning_rate": 9.994086463376534e-06,
+      "loss": 0.9039,
+      "step": 140
+    },
+    {
+      "epoch": 0.3434529285495247,
+      "eval_loss": 0.9195936918258667,
+      "eval_runtime": 3.2219,
+      "eval_samples_per_second": 24.52,
+      "eval_steps_per_second": 6.208,
+      "step": 140
+    },
+    {
+      "epoch": 0.36798528058877644,
+      "grad_norm": 0.8446605004620666,
+      "learning_rate": 9.985087632242634e-06,
+      "loss": 0.8916,
+      "step": 150
+    },
+    {
+      "epoch": 0.3925176326280282,
+      "grad_norm": 0.7548473430154016,
+      "learning_rate": 9.972008067598492e-06,
+      "loss": 0.8925,
+      "step": 160
+    },
+    {
+      "epoch": 0.3925176326280282,
+      "eval_loss": 0.9043188095092773,
+      "eval_runtime": 3.2476,
+      "eval_samples_per_second": 24.325,
+      "eval_steps_per_second": 6.158,
+      "step": 160
+    },
+    {
+      "epoch": 0.41704998466728,
+      "grad_norm": 0.6914000510528906,
+      "learning_rate": 9.954858476220331e-06,
+      "loss": 0.8859,
+      "step": 170
+    },
+    {
+      "epoch": 0.44158233670653174,
+      "grad_norm": 0.6279329539612621,
+      "learning_rate": 9.933652896559907e-06,
+      "loss": 0.8845,
+      "step": 180
+    },
+    {
+      "epoch": 0.44158233670653174,
+      "eval_loss": 0.8936107754707336,
+      "eval_runtime": 3.2343,
+      "eval_samples_per_second": 24.426,
+      "eval_steps_per_second": 6.184,
+      "step": 180
+    },
+    {
+      "epoch": 0.4661146887457835,
+      "grad_norm": 0.747923448105748,
+      "learning_rate": 9.9084086872528e-06,
+      "loss": 0.8708,
+      "step": 190
+    },
+    {
+      "epoch": 0.49064704078503524,
+      "grad_norm": 0.6810626276029981,
+      "learning_rate": 9.879146512908832e-06,
+      "loss": 0.8718,
+      "step": 200
+    },
+    {
+      "epoch": 0.49064704078503524,
+      "eval_loss": 0.8834576606750488,
+      "eval_runtime": 3.2255,
+      "eval_samples_per_second": 24.493,
+      "eval_steps_per_second": 6.201,
+      "step": 200
+    },
+    {
+      "epoch": 0.515179392824287,
+      "grad_norm": 0.7788033592184153,
+      "learning_rate": 9.845890327196269e-06,
+      "loss": 0.8601,
+      "step": 210
+    },
+    {
+      "epoch": 0.5397117448635388,
+      "grad_norm": 0.7063664080670593,
+      "learning_rate": 9.808667353233637e-06,
+      "loss": 0.866,
+      "step": 220
+    },
+    {
+      "epoch": 0.5397117448635388,
+      "eval_loss": 0.8751548528671265,
+      "eval_runtime": 3.2323,
+      "eval_samples_per_second": 24.441,
+      "eval_steps_per_second": 6.188,
+      "step": 220
+    },
+    {
+      "epoch": 0.5642440969027905,
+      "grad_norm": 0.6941671762903114,
+      "learning_rate": 9.767508061305188e-06,
+      "loss": 0.8589,
+      "step": 230
+    },
+    {
+      "epoch": 0.5887764489420423,
+      "grad_norm": 0.6763821128552876,
+      "learning_rate": 9.722446143918307e-06,
+      "loss": 0.8475,
+      "step": 240
+    },
+    {
+      "epoch": 0.5887764489420423,
+      "eval_loss": 0.8678768277168274,
+      "eval_runtime": 3.2586,
+      "eval_samples_per_second": 24.244,
+      "eval_steps_per_second": 6.138,
+      "step": 240
+    },
+    {
+      "epoch": 0.613308800981294,
+      "grad_norm": 0.6434953779579147,
+      "learning_rate": 9.673518488223203e-06,
+      "loss": 0.8385,
+      "step": 250
+    },
+    {
+      "epoch": 0.6378411530205459,
+      "grad_norm": 0.6414335352009913,
+      "learning_rate": 9.62076514581754e-06,
+      "loss": 0.8429,
+      "step": 260
+    },
+    {
+      "epoch": 0.6378411530205459,
+      "eval_loss": 0.8620171546936035,
+      "eval_runtime": 3.2458,
+      "eval_samples_per_second": 24.339,
+      "eval_steps_per_second": 6.162,
+      "step": 260
+    },
+    {
+      "epoch": 0.6623735050597976,
+      "grad_norm": 0.6177346684876659,
+      "learning_rate": 9.564229299960679e-06,
+      "loss": 0.8391,
+      "step": 270
+    },
+    {
+      "epoch": 0.6869058570990494,
+      "grad_norm": 0.7021878424452098,
+      "learning_rate": 9.503957230224356e-06,
+      "loss": 0.8396,
+      "step": 280
+    },
+    {
+      "epoch": 0.6869058570990494,
+      "eval_loss": 0.8567876815795898,
+      "eval_runtime": 3.2301,
+      "eval_samples_per_second": 24.457,
+      "eval_steps_per_second": 6.192,
+      "step": 280
+    },
+    {
+      "epoch": 0.7114382091383011,
+      "grad_norm": 0.6774899531719633,
+      "learning_rate": 9.439998274608797e-06,
+      "loss": 0.8369,
+      "step": 290
+    },
+    {
+      "epoch": 0.7359705611775529,
+      "grad_norm": 0.6200422662890627,
+      "learning_rate": 9.372404789155199e-06,
+      "loss": 0.8365,
+      "step": 300
+    },
+    {
+      "epoch": 0.7359705611775529,
+      "eval_loss": 0.8507375717163086,
+      "eval_runtime": 3.2264,
+      "eval_samples_per_second": 24.485,
+      "eval_steps_per_second": 6.199,
+      "step": 300
+    },
+    {
+      "epoch": 0.7605029132168046,
+      "grad_norm": 0.6083768903694923,
+      "learning_rate": 9.30123210508771e-06,
+      "loss": 0.8323,
+      "step": 310
+    },
+    {
+      "epoch": 0.7850352652560564,
+      "grad_norm": 0.6276586503575396,
+      "learning_rate": 9.226538483519938e-06,
+      "loss": 0.8284,
+      "step": 320
+    },
+    {
+      "epoch": 0.7850352652560564,
+      "eval_loss": 0.8456504940986633,
+      "eval_runtime": 3.2442,
+      "eval_samples_per_second": 24.351,
+      "eval_steps_per_second": 6.165,
+      "step": 320
+    },
+    {
+      "epoch": 0.8095676172953082,
+      "grad_norm": 0.6135798006562337,
+      "learning_rate": 9.148385067763094e-06,
+      "loss": 0.8224,
+      "step": 330
+    },
+    {
+      "epoch": 0.83409996933456,
+      "grad_norm": 0.6858590248193231,
+      "learning_rate": 9.066835833274818e-06,
+      "loss": 0.8314,
+      "step": 340
+    },
+    {
+      "epoch": 0.83409996933456,
+      "eval_loss": 0.8415058851242065,
+      "eval_runtime": 3.2608,
+      "eval_samples_per_second": 24.227,
+      "eval_steps_per_second": 6.133,
+      "step": 340
+    },
+    {
+      "epoch": 0.8586323213738117,
+      "grad_norm": 0.6653290649299476,
+      "learning_rate": 8.981957535289612e-06,
+      "loss": 0.8248,
+      "step": 350
+    },
+    {
+      "epoch": 0.8831646734130635,
+      "grad_norm": 0.653483902265936,
+      "learning_rate": 8.893819654173804e-06,
+      "loss": 0.828,
+      "step": 360
+    },
+    {
+      "epoch": 0.8831646734130635,
+      "eval_loss": 0.8383563756942749,
+      "eval_runtime": 3.2339,
+      "eval_samples_per_second": 24.429,
+      "eval_steps_per_second": 6.185,
+      "step": 360
+    },
+    {
+      "epoch": 0.9076970254523152,
+      "grad_norm": 0.6182165779851636,
+      "learning_rate": 8.802494338549753e-06,
+      "loss": 0.8218,
+      "step": 370
+    },
+    {
+      "epoch": 0.932229377491567,
+      "grad_norm": 0.620725102354971,
+      "learning_rate": 8.708056346235834e-06,
+      "loss": 0.8084,
+      "step": 380
+    },
+    {
+      "epoch": 0.932229377491567,
+      "eval_loss": 0.8347155451774597,
+      "eval_runtime": 3.2259,
+      "eval_samples_per_second": 24.489,
+      "eval_steps_per_second": 6.2,
+      "step": 380
+    },
+    {
+      "epoch": 0.9567617295308187,
+      "grad_norm": 0.6134856665584789,
+      "learning_rate": 8.610582983050581e-06,
+      "loss": 0.826,
+      "step": 390
+    },
+    {
+      "epoch": 0.9812940815700705,
+      "grad_norm": 0.6151238147796793,
+      "learning_rate": 8.510154039531063e-06,
+      "loss": 0.8217,
+      "step": 400
+    },
+    {
+      "epoch": 0.9812940815700705,
+      "eval_loss": 0.8309343457221985,
+      "eval_runtime": 3.28,
+      "eval_samples_per_second": 24.085,
+      "eval_steps_per_second": 6.098,
+      "step": 400
+    },
+    {
+      "epoch": 1.006133088009813,
+      "grad_norm": 0.6567379397981333,
+      "learning_rate": 8.406851725617295e-06,
+      "loss": 0.8475,
+      "step": 410
+    },
+    {
+      "epoch": 1.0306654400490647,
+      "grad_norm": 0.6438546869143422,
+      "learning_rate": 8.30076060335616e-06,
+      "loss": 0.778,
+      "step": 420
+    },
+    {
+      "epoch": 1.0306654400490647,
+      "eval_loss": 0.8290137052536011,
+      "eval_runtime": 3.2252,
+      "eval_samples_per_second": 24.494,
+      "eval_steps_per_second": 6.201,
+      "step": 420
+    },
+    {
+      "epoch": 1.0551977920883164,
+      "grad_norm": 0.5947085559389068,
+      "learning_rate": 8.19196751767992e-06,
+      "loss": 0.7817,
+      "step": 430
+    },
+    {
+      "epoch": 1.0797301441275682,
+      "grad_norm": 0.6577116987458262,
+      "learning_rate": 8.08056152531599e-06,
+      "loss": 0.7756,
+      "step": 440
+    },
+    {
+      "epoch": 1.0797301441275682,
+      "eval_loss": 0.8259589076042175,
+      "eval_runtime": 3.2219,
+      "eval_samples_per_second": 24.52,
+      "eval_steps_per_second": 6.207,
+      "step": 440
+    },
+    {
+      "epoch": 1.10426249616682,
+      "grad_norm": 0.647328944123465,
+      "learning_rate": 7.96663382188616e-06,
+      "loss": 0.7736,
+      "step": 450
+    },
+    {
+      "epoch": 1.1287948482060717,
+      "grad_norm": 0.6317678678629681,
+      "learning_rate": 7.850277667254931e-06,
+      "loss": 0.7709,
+      "step": 460
+    },
+    {
+      "epoch": 1.1287948482060717,
+      "eval_loss": 0.8239238858222961,
+      "eval_runtime": 3.2695,
+      "eval_samples_per_second": 24.163,
+      "eval_steps_per_second": 6.117,
+      "step": 460
+    },
+    {
+      "epoch": 1.1533272002453234,
+      "grad_norm": 0.6994398660973125,
+      "learning_rate": 7.731588309188113e-06,
+      "loss": 0.7785,
+      "step": 470
+    },
+    {
+      "epoch": 1.1778595522845752,
+      "grad_norm": 0.6724295151080316,
+      "learning_rate": 7.6106629053841255e-06,
+      "loss": 0.7762,
+      "step": 480
+    },
+    {
+      "epoch": 1.1778595522845752,
+      "eval_loss": 0.8216440081596375,
+      "eval_runtime": 3.2207,
+      "eval_samples_per_second": 24.529,
+      "eval_steps_per_second": 6.21,
+      "step": 480
+    },
+    {
+      "epoch": 1.2023919043238271,
+      "grad_norm": 0.6320248080962604,
+      "learning_rate": 7.487600443941863e-06,
+      "loss": 0.7723,
+      "step": 490
+    },
+    {
+      "epoch": 1.226924256363079,
+      "grad_norm": 0.6217306847757799,
+      "learning_rate": 7.3625016623302155e-06,
+      "loss": 0.7804,
+      "step": 500
+    },
+    {
+      "epoch": 1.226924256363079,
+      "eval_loss": 0.8192575573921204,
+      "eval_runtime": 3.2358,
+      "eval_samples_per_second": 24.414,
+      "eval_steps_per_second": 6.181,
+      "step": 500
+    },
+    {
+      "epoch": 1.2514566084023306,
+      "grad_norm": 0.6200515640118288,
+      "learning_rate": 7.235468964925571e-06,
+      "loss": 0.7789,
+      "step": 510
+    },
+    {
+      "epoch": 1.2759889604415824,
+      "grad_norm": 0.6343301876087556,
+      "learning_rate": 7.106606339184805e-06,
+      "loss": 0.772,
+      "step": 520
+    },
+    {
+      "epoch": 1.2759889604415824,
+      "eval_loss": 0.8179527521133423,
+      "eval_runtime": 3.2613,
+      "eval_samples_per_second": 24.224,
+      "eval_steps_per_second": 6.133,
+      "step": 520
+    },
+    {
+      "epoch": 1.3005213124808341,
+      "grad_norm": 0.6328736580426868,
+      "learning_rate": 6.976019270522394e-06,
+      "loss": 0.7785,
+      "step": 530
+    },
+    {
+      "epoch": 1.3250536645200859,
+      "grad_norm": 0.6385372493592535,
+      "learning_rate": 6.843814655961302e-06,
+      "loss": 0.7711,
+      "step": 540
+    },
+    {
+      "epoch": 1.3250536645200859,
+      "eval_loss": 0.814644992351532,
+      "eval_runtime": 3.2223,
+      "eval_samples_per_second": 24.517,
+      "eval_steps_per_second": 6.207,
+      "step": 540
+    },
+    {
+      "epoch": 1.3495860165593376,
+      "grad_norm": 0.609021186816327,
+      "learning_rate": 6.710100716628345e-06,
+      "loss": 0.7669,
+      "step": 550
+    },
+    {
+      "epoch": 1.3741183685985894,
+      "grad_norm": 0.6424751080031044,
+      "learning_rate": 6.574986909165668e-06,
+      "loss": 0.7701,
+      "step": 560
+    },
+    {
+      "epoch": 1.3741183685985894,
+      "eval_loss": 0.813719630241394,
+      "eval_runtime": 3.2505,
+      "eval_samples_per_second": 24.304,
+      "eval_steps_per_second": 6.153,
+      "step": 560
+    },
+    {
+      "epoch": 1.3986507206378411,
+      "grad_norm": 0.6364206410592088,
+      "learning_rate": 6.438583836130835e-06,
+      "loss": 0.7779,
+      "step": 570
+    },
+    {
+      "epoch": 1.4231830726770929,
+      "grad_norm": 0.5949610647626281,
+      "learning_rate": 6.301003155458894e-06,
+      "loss": 0.7663,
+      "step": 580
+    },
+    {
+      "epoch": 1.4231830726770929,
+      "eval_loss": 0.8118559122085571,
+      "eval_runtime": 3.2465,
+      "eval_samples_per_second": 24.334,
+      "eval_steps_per_second": 6.161,
+      "step": 580
+    },
+    {
+      "epoch": 1.4477154247163446,
+      "grad_norm": 0.5980152115219746,
+      "learning_rate": 6.162357489060508e-06,
+      "loss": 0.7658,
+      "step": 590
+    },
+    {
+      "epoch": 1.4722477767555964,
+      "grad_norm": 0.613551293850175,
+      "learning_rate": 6.022760330631006e-06,
+      "loss": 0.7623,
+      "step": 600
+    },
+    {
+      "epoch": 1.4722477767555964,
+      "eval_loss": 0.8094308972358704,
+      "eval_runtime": 3.2741,
+      "eval_samples_per_second": 24.129,
+      "eval_steps_per_second": 6.109,
+      "step": 600
+    },
+    {
+      "epoch": 1.4967801287948483,
+      "grad_norm": 0.5929932888694315,
+      "learning_rate": 5.8823259527457885e-06,
+      "loss": 0.7645,
+      "step": 610
+    },
+    {
+      "epoch": 1.5213124808341,
+      "grad_norm": 0.6173688720575873,
+      "learning_rate": 5.74116931331816e-06,
+      "loss": 0.7681,
+      "step": 620
+    },
+    {
+      "epoch": 1.5213124808341,
+      "eval_loss": 0.807237982749939,
+      "eval_runtime": 3.3481,
+      "eval_samples_per_second": 23.595,
+      "eval_steps_per_second": 5.974,
+      "step": 620
+    },
+    {
+      "epoch": 1.5458448328733518,
+      "grad_norm": 0.5869452620685718,
+      "learning_rate": 5.599405961496137e-06,
+      "loss": 0.7571,
+      "step": 630
+    },
+    {
+      "epoch": 1.5703771849126036,
+      "grad_norm": 0.5891167602159463,
+      "learning_rate": 5.457151943075297e-06,
+      "loss": 0.7677,
+      "step": 640
+    },
+    {
+      "epoch": 1.5703771849126036,
+      "eval_loss": 0.8063360452651978,
+      "eval_runtime": 3.2241,
+      "eval_samples_per_second": 24.503,
+      "eval_steps_per_second": 6.203,
+      "step": 640
+    },
+    {
+      "epoch": 1.5949095369518553,
+      "grad_norm": 0.5975322420438561,
+      "learning_rate": 5.314523705505066e-06,
+      "loss": 0.7596,
+      "step": 650
+    },
+    {
+      "epoch": 1.619441888991107,
+      "grad_norm": 0.620029726002314,
+      "learning_rate": 5.171638002566218e-06,
+      "loss": 0.7643,
+      "step": 660
+    },
+    {
+      "epoch": 1.619441888991107,
+      "eval_loss": 0.804570198059082,
+      "eval_runtime": 3.2384,
+      "eval_samples_per_second": 24.395,
+      "eval_steps_per_second": 6.176,
+      "step": 660
+    },
+    {
+      "epoch": 1.6439742410303588,
+      "grad_norm": 0.6262842299013245,
+      "learning_rate": 5.0286117987976215e-06,
+      "loss": 0.7609,
+      "step": 670
+    },
+    {
+      "epoch": 1.6685065930696106,
+      "grad_norm": 0.6411001339949628,
+      "learning_rate": 4.885562173750452e-06,
+      "loss": 0.7618,
+      "step": 680
+    },
+    {
+      "epoch": 1.6685065930696106,
+      "eval_loss": 0.8032476305961609,
+      "eval_runtime": 3.236,
+      "eval_samples_per_second": 24.413,
+      "eval_steps_per_second": 6.18,
+      "step": 680
+    },
+    {
+      "epoch": 1.6930389451088623,
+      "grad_norm": 0.5916445759107795,
+      "learning_rate": 4.742606226148268e-06,
+      "loss": 0.7614,
+      "step": 690
+    },
+    {
+      "epoch": 1.717571297148114,
+      "grad_norm": 0.5926125741058034,
+      "learning_rate": 4.599860978031383e-06,
+      "loss": 0.7665,
+      "step": 700
+    },
+    {
+      "epoch": 1.717571297148114,
+      "eval_loss": 0.8017654418945312,
+      "eval_runtime": 3.2431,
+      "eval_samples_per_second": 24.36,
+      "eval_steps_per_second": 6.167,
+      "step": 700
+    },
+    {
+      "epoch": 1.7421036491873658,
+      "grad_norm": 0.6038286337474377,
+      "learning_rate": 4.457443278964016e-06,
+      "loss": 0.7679,
+      "step": 710
+    },
+    {
+      "epoch": 1.7666360012266176,
+      "grad_norm": 0.592330275454734,
+      "learning_rate": 4.3154697103826235e-06,
+      "loss": 0.7614,
+      "step": 720
+    },
+    {
+      "epoch": 1.7666360012266176,
+      "eval_loss": 0.8002952337265015,
+      "eval_runtime": 3.2255,
+      "eval_samples_per_second": 24.492,
+      "eval_steps_per_second": 6.201,
+      "step": 720
+    },
+    {
+      "epoch": 1.7911683532658693,
+      "grad_norm": 0.546004408597974,
+      "learning_rate": 4.174056490163727e-06,
+      "loss": 0.7545,
+      "step": 730
+    },
+    {
+      "epoch": 1.815700705305121,
+      "grad_norm": 0.5744374392780872,
+      "learning_rate": 4.03331937748933e-06,
+      "loss": 0.7557,
+      "step": 740
+    },
+    {
+      "epoch": 1.815700705305121,
+      "eval_loss": 0.7996227145195007,
+      "eval_runtime": 3.3245,
+      "eval_samples_per_second": 23.763,
+      "eval_steps_per_second": 6.016,
+      "step": 740
+    },
+    {
+      "epoch": 1.8402330573443728,
+      "grad_norm": 0.60707319276783,
+      "learning_rate": 3.893373578087838e-06,
+      "loss": 0.7527,
+      "step": 750
+    },
+    {
+      "epoch": 1.8647654093836246,
+      "grad_norm": 0.5721720524185794,
+      "learning_rate": 3.754333649928007e-06,
+      "loss": 0.7534,
+      "step": 760
+    },
+    {
+      "epoch": 1.8647654093836246,
+      "eval_loss": 0.7987948656082153,
+      "eval_runtime": 3.2466,
+      "eval_samples_per_second": 24.333,
+      "eval_steps_per_second": 6.16,
+      "step": 760
+    },
+    {
+      "epoch": 1.8892977614228763,
+      "grad_norm": 0.558280936125822,
+      "learning_rate": 3.616313409443163e-06,
+      "loss": 0.756,
+      "step": 770
+    },
+    {
+      "epoch": 1.913830113462128,
+      "grad_norm": 0.573221328629841,
+      "learning_rate": 3.4794258383624115e-06,
+      "loss": 0.7464,
+      "step": 780
+    },
+    {
+      "epoch": 1.913830113462128,
+      "eval_loss": 0.7975552678108215,
+      "eval_runtime": 3.2695,
+      "eval_samples_per_second": 24.163,
+      "eval_steps_per_second": 6.117,
+      "step": 780
+    },
+    {
+      "epoch": 1.9383624655013798,
+      "grad_norm": 0.5571524597964522,
+      "learning_rate": 3.34378299122514e-06,
+      "loss": 0.7603,
+      "step": 790
+    },
+    {
+      "epoch": 1.9628948175406316,
+      "grad_norm": 0.5701811044528794,
+      "learning_rate": 3.2094959036545057e-06,
+      "loss": 0.7626,
+      "step": 800
+    },
+    {
+      "epoch": 1.9628948175406316,
+      "eval_loss": 0.7968281507492065,
+      "eval_runtime": 3.2581,
+      "eval_samples_per_second": 24.247,
+      "eval_steps_per_second": 6.138,
+      "step": 800
+    },
+    {
+      "epoch": 1.9874271695798835,
+      "grad_norm": 0.5927397235414454,
+      "learning_rate": 3.076674501464994e-06,
+      "loss": 0.7597,
+      "step": 810
+    },
+    {
+      "epoch": 2.012266176019626,
+      "grad_norm": 0.593400171337047,
+      "learning_rate": 2.9454275106784457e-06,
+      "loss": 0.7849,
+      "step": 820
+    },
+    {
+      "epoch": 2.012266176019626,
+      "eval_loss": 0.7972843050956726,
+      "eval_runtime": 3.2399,
+      "eval_samples_per_second": 24.384,
+      "eval_steps_per_second": 6.173,
+      "step": 820
+    },
+    {
+      "epoch": 2.0367985280588776,
+      "grad_norm": 0.561241980320335,
+      "learning_rate": 2.815862368522236e-06,
+      "loss": 0.7286,
+      "step": 830
+    },
+    {
+      "epoch": 2.0613308800981294,
+      "grad_norm": 0.5894246337251206,
+      "learning_rate": 2.6880851354824277e-06,
+      "loss": 0.7264,
+      "step": 840
+    },
+    {
+      "epoch": 2.0613308800981294,
+      "eval_loss": 0.7973536252975464,
+      "eval_runtime": 3.2137,
+      "eval_samples_per_second": 24.582,
+      "eval_steps_per_second": 6.223,
+      "step": 840
+    },
+    {
+      "epoch": 2.085863232137381,
+      "grad_norm": 0.5805134102436514,
+      "learning_rate": 2.5622004084839248e-06,
+      "loss": 0.7281,
+      "step": 850
+    },
+    {
+      "epoch": 2.110395584176633,
+      "grad_norm": 0.5465453480702682,
+      "learning_rate": 2.438311235268674e-06,
+      "loss": 0.7299,
+      "step": 860
+    },
+    {
+      "epoch": 2.110395584176633,
+      "eval_loss": 0.7964964509010315,
+      "eval_runtime": 3.2447,
+      "eval_samples_per_second": 24.348,
+      "eval_steps_per_second": 6.164,
+      "step": 860
+    },
+    {
+      "epoch": 2.1349279362158846,
+      "grad_norm": 0.5711000673693014,
+      "learning_rate": 2.316519030041998e-06,
+      "loss": 0.7301,
+      "step": 870
+    },
+    {
+      "epoch": 2.1594602882551364,
+      "grad_norm": 0.5576952609996458,
+      "learning_rate": 2.1969234904561386e-06,
+      "loss": 0.7251,
+      "step": 880
+    },
+    {
+      "epoch": 2.1594602882551364,
+      "eval_loss": 0.7965396642684937,
+      "eval_runtime": 3.2403,
+      "eval_samples_per_second": 24.381,
+      "eval_steps_per_second": 6.172,
+      "step": 880
+    },
+    {
+      "epoch": 2.183992640294388,
+      "grad_norm": 0.536356553752016,
+      "learning_rate": 2.0796225159989403e-06,
+      "loss": 0.7245,
+      "step": 890
+    },
+    {
+      "epoch": 2.20852499233364,
+      "grad_norm": 0.5563367480900646,
+      "learning_rate": 1.9647121278544996e-06,
+      "loss": 0.7271,
+      "step": 900
+    },
+    {
+      "epoch": 2.20852499233364,
+      "eval_loss": 0.7961751818656921,
+      "eval_runtime": 3.2421,
+      "eval_samples_per_second": 24.367,
+      "eval_steps_per_second": 6.169,
+      "step": 900
+    },
+    {
+      "epoch": 2.2330573443728916,
+      "grad_norm": 0.5693189868816336,
+      "learning_rate": 1.8522863903013504e-06,
+      "loss": 0.7321,
+      "step": 910
+    },
+    {
+      "epoch": 2.2575896964121434,
+      "grad_norm": 0.5735126257618532,
+      "learning_rate": 1.742437333712577e-06,
+      "loss": 0.7317,
+      "step": 920
+    },
+    {
+      "epoch": 2.2575896964121434,
+      "eval_loss": 0.7955650091171265,
+      "eval_runtime": 3.2252,
+      "eval_samples_per_second": 24.494,
+      "eval_steps_per_second": 6.201,
+      "step": 920
+    },
+    {
+      "epoch": 2.282122048451395,
+      "grad_norm": 0.6011215391129776,
+      "learning_rate": 1.6352548792208356e-06,
+      "loss": 0.7295,
+      "step": 930
+    },
+    {
+      "epoch": 2.306654400490647,
+      "grad_norm": 0.5570940285155812,
+      "learning_rate": 1.5308267651099917e-06,
+      "loss": 0.7243,
+      "step": 940
+    },
+    {
+      "epoch": 2.306654400490647,
+      "eval_loss": 0.7952836155891418,
+      "eval_runtime": 3.2635,
+      "eval_samples_per_second": 24.207,
+      "eval_steps_per_second": 6.128,
+      "step": 940
+    },
+    {
+      "epoch": 2.3311867525298986,
+      "grad_norm": 0.5410445515616614,
+      "learning_rate": 1.4292384749935928e-06,
+      "loss": 0.7295,
+      "step": 950
+    },
+    {
+      "epoch": 2.3557191045691503,
+      "grad_norm": 0.5595739216163256,
+      "learning_rate": 1.330573167839005e-06,
+      "loss": 0.7216,
+      "step": 960
+    },
+    {
+      "epoch": 2.3557191045691503,
+      "eval_loss": 0.7947081923484802,
+      "eval_runtime": 3.25,
+      "eval_samples_per_second": 24.308,
+      "eval_steps_per_second": 6.154,
+      "step": 960
+    },
+    {
+      "epoch": 2.380251456608402,
+      "grad_norm": 0.532819336636009,
+      "learning_rate": 1.234911609894467e-06,
+      "loss": 0.7221,
+      "step": 970
+    },
+    {
+      "epoch": 2.4047838086476543,
+      "grad_norm": 0.543586019360024,
+      "learning_rate": 1.142332108574805e-06,
+      "loss": 0.7273,
+      "step": 980
+    },
+    {
+      "epoch": 2.4047838086476543,
+      "eval_loss": 0.7941427230834961,
+      "eval_runtime": 3.2573,
+      "eval_samples_per_second": 24.253,
+      "eval_steps_per_second": 6.14,
+      "step": 980
+    },
+    {
+      "epoch": 2.429316160686906,
+      "grad_norm": 0.5577945748915079,
+      "learning_rate": 1.0529104483599023e-06,
+      "loss": 0.7289,
+      "step": 990
+    },
+    {
+      "epoch": 2.453848512726158,
+      "grad_norm": 0.5409696592334879,
+      "learning_rate": 9.667198287584345e-07,
+      "loss": 0.7185,
+      "step": 1000
+    },
+    {
+      "epoch": 2.453848512726158,
+      "eval_loss": 0.7937620878219604,
+      "eval_runtime": 3.2083,
+      "eval_samples_per_second": 24.624,
+      "eval_steps_per_second": 6.234,
+      "step": 1000
+    },
+    {
+      "epoch": 2.4783808647654095,
+      "grad_norm": 0.55237204059885,
+      "learning_rate": 8.838308043876154e-07,
+      "loss": 0.7221,
+      "step": 1010
+    },
+    {
+      "epoch": 2.5029132168046613,
+      "grad_norm": 0.5346685035163194,
+      "learning_rate": 8.043112272180309e-07,
+      "loss": 0.7236,
+      "step": 1020
+    },
+    {
+      "epoch": 2.5029132168046613,
+      "eval_loss": 0.7935998439788818,
+      "eval_runtime": 3.234,
+      "eval_samples_per_second": 24.428,
+      "eval_steps_per_second": 6.184,
+      "step": 1020
+    },
+    {
+      "epoch": 2.527445568843913,
+      "grad_norm": 0.5372160722510528,
+      "learning_rate": 7.282261910308163e-07,
+      "loss": 0.7316,
+      "step": 1030
+    },
+    {
+      "epoch": 2.5519779208831648,
+      "grad_norm": 0.5306493562577365,
+      "learning_rate": 6.556379781326633e-07,
+      "loss": 0.7247,
+      "step": 1040
+    },
+    {
+      "epoch": 2.5519779208831648,
+      "eval_loss": 0.7934552431106567,
+      "eval_runtime": 3.2327,
+      "eval_samples_per_second": 24.438,
+      "eval_steps_per_second": 6.187,
+      "step": 1040
+    },
+    {
+      "epoch": 2.5765102729224165,
+      "grad_norm": 0.5178178514238788,
+      "learning_rate": 5.866060083722624e-07,
+      "loss": 0.7258,
+      "step": 1050
+    },
+    {
+      "epoch": 2.6010426249616683,
+      "grad_norm": 0.5462317156108272,
+      "learning_rate": 5.211867904999258e-07,
+      "loss": 0.727,
+      "step": 1060
+    },
+    {
+      "epoch": 2.6010426249616683,
+      "eval_loss": 0.7931656837463379,
+      "eval_runtime": 3.2352,
+      "eval_samples_per_second": 24.419,
+      "eval_steps_per_second": 6.182,
+      "step": 1060
+    },
+    {
+      "epoch": 2.62557497700092,
+      "grad_norm": 0.5220579343155467,
+      "learning_rate": 4.594338759101852e-07,
+      "loss": 0.725,
+      "step": 1070
+    },
+    {
+      "epoch": 2.6501073290401718,
+      "grad_norm": 0.5312213886454074,
+      "learning_rate": 4.0139781480526776e-07,
+      "loss": 0.7209,
+      "step": 1080
+    },
+    {
+      "epoch": 2.6501073290401718,
+      "eval_loss": 0.7929045557975769,
+      "eval_runtime": 3.222,
+      "eval_samples_per_second": 24.519,
+      "eval_steps_per_second": 6.207,
+      "step": 1080
+    },
+    {
+      "epoch": 2.6746396810794235,
+      "grad_norm": 0.5465813970697985,
+      "learning_rate": 3.4712611481529266e-07,
+      "loss": 0.7284,
+      "step": 1090
+    },
+    {
+      "epoch": 2.6991720331186753,
+      "grad_norm": 0.5122023913210224,
+      "learning_rate": 2.9666320210909703e-07,
+      "loss": 0.7225,
+      "step": 1100
+    },
+    {
+      "epoch": 2.6991720331186753,
+      "eval_loss": 0.7927907705307007,
+      "eval_runtime": 3.2231,
+      "eval_samples_per_second": 24.51,
+      "eval_steps_per_second": 6.205,
+      "step": 1100
+    },
+    {
+      "epoch": 2.723704385157927,
+      "grad_norm": 0.533599198086878,
+      "learning_rate": 2.500503850274949e-07,
+      "loss": 0.7293,
+      "step": 1110
+    },
+    {
+      "epoch": 2.7482367371971788,
+      "grad_norm": 0.5340176542148305,
+      "learning_rate": 2.0732582026876448e-07,
+      "loss": 0.722,
+      "step": 1120
+    },
+    {
+      "epoch": 2.7482367371971788,
+      "eval_loss": 0.7927077412605286,
+      "eval_runtime": 3.2381,
+      "eval_samples_per_second": 24.397,
+      "eval_steps_per_second": 6.176,
+      "step": 1120
+    },
+    {
+      "epoch": 2.7727690892364305,
+      "grad_norm": 0.5341232058197358,
+      "learning_rate": 1.6852448165402546e-07,
+      "loss": 0.7186,
+      "step": 1130
+    },
+    {
+      "epoch": 2.7973014412756823,
+      "grad_norm": 0.555997817609322,
+      "learning_rate": 1.3367813149808728e-07,
+      "loss": 0.7276,
+      "step": 1140
+    },
+    {
+      "epoch": 2.7973014412756823,
+      "eval_loss": 0.7926952242851257,
+      "eval_runtime": 3.2267,
+      "eval_samples_per_second": 24.483,
+      "eval_steps_per_second": 6.198,
+      "step": 1140
+    },
+    {
+      "epoch": 2.821833793314934,
+      "grad_norm": 0.5419789908629483,
+      "learning_rate": 1.0281529460919592e-07,
+      "loss": 0.7205,
+      "step": 1150
+    },
+    {
+      "epoch": 2.8463661453541858,
+      "grad_norm": 0.5465053730798224,
+      "learning_rate": 7.59612349389599e-08,
+      "loss": 0.7249,
+      "step": 1160
+    },
+    {
+      "epoch": 2.8463661453541858,
+      "eval_loss": 0.7926090955734253,
+      "eval_runtime": 3.2422,
+      "eval_samples_per_second": 24.367,
+      "eval_steps_per_second": 6.169,
+      "step": 1160
+    },
+    {
+      "epoch": 2.8708984973934375,
+      "grad_norm": 0.532815905161259,
+      "learning_rate": 5.3137934901585364e-08,
+      "loss": 0.721,
+      "step": 1170
+    },
+    {
+      "epoch": 2.8954308494326892,
+      "grad_norm": 0.5438211558294759,
+      "learning_rate": 3.4364077379329674e-08,
+      "loss": 0.7314,
+      "step": 1180
+    },
+    {
+      "epoch": 2.8954308494326892,
+      "eval_loss": 0.7926509380340576,
+      "eval_runtime": 3.2304,
+      "eval_samples_per_second": 24.455,
+      "eval_steps_per_second": 6.191,
+      "step": 1180
+    },
+    {
+      "epoch": 2.919963201471941,
+      "grad_norm": 0.5406808251158669,
+      "learning_rate": 1.9655030428917254e-08,
+      "loss": 0.7292,
+      "step": 1190
+    },
+    {
+      "epoch": 2.9444955535111927,
+      "grad_norm": 0.5341446204049535,
+      "learning_rate": 9.022834701429839e-09,
+      "loss": 0.7247,
+      "step": 1200
+    },
+    {
+      "epoch": 2.9444955535111927,
+      "eval_loss": 0.7925666570663452,
+      "eval_runtime": 3.2369,
+      "eval_samples_per_second": 24.406,
+      "eval_steps_per_second": 6.179,
+      "step": 1200
+    },
+    {
+      "epoch": 2.9690279055504445,
+      "grad_norm": 0.544093441465175,
+      "learning_rate": 2.4761935859746178e-09,
+      "loss": 0.7291,
+      "step": 1210
+    },
+    {
+      "epoch": 2.9935602575896967,
+      "grad_norm": 0.5385799983177132,
+      "learning_rate": 2.046608519046256e-11,
+      "loss": 0.7248,
+      "step": 1220
+    },
+    {
+      "epoch": 2.9935602575896967,
+      "eval_loss": 0.7926931977272034,
+      "eval_runtime": 3.2282,
+      "eval_samples_per_second": 24.472,
+      "eval_steps_per_second": 6.195,
+      "step": 1220
+    },
+    {
+      "epoch": 2.9960134927936215,
+      "step": 1221,
+      "total_flos": 337549181616128.0,
+      "train_loss": 0.809739662817909,
+      "train_runtime": 19632.9853,
+      "train_samples_per_second": 5.978,
+      "train_steps_per_second": 0.062
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1221,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 337549181616128.0,
+  "train_batch_size": 3,
+  "trial_name": null,
+  "trial_params": null
+}

scenario_cot_ratio_1.5B/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d83aa29ed01dbe6f9b910659d5df666e1d6641e0d9f28f0317494f8076df83d
+size 7288