Model save

.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

README.md

@@ -0,0 +1,68 @@
+---
+base_model: Qwen/Qwen2.5-1.5B-Instruct
+library_name: transformers
+model_name: Qwen2.5-1.5B-Open-R1-GRPO
+tags:
+- generated_from_trainer
+- trl
+- grpo
+licence: license
+---
+
+# Model Card for Qwen2.5-1.5B-Open-R1-GRPO
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="shuheikurita/Qwen2.5-1.5B-Open-R1-GRPO", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/shuheikurita/huggingface/runs/lxcgwgxl) 
+
+
+This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
+
+### Framework versions
+
+- TRL: 0.16.0.dev0
+- Transformers: 4.49.0.dev0
+- Pytorch: 2.5.1
+- Datasets: 3.3.0
+- Tokenizers: 0.21.0
+
+## Citations
+
+Cite GRPO as:
+
+```bibtex
+@article{zhihong2024deepseekmath,
+    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
+    author       = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
+    year         = 2024,
+    eprint       = {arXiv:2402.03300},
+}
+
+```
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

added_tokens.json

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

all_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 0.0,
+    "train_loss": 0.017702588886127465,
+    "train_runtime": 38308.166,
+    "train_samples": 72441,
+    "train_samples_per_second": 1.891,
+    "train_steps_per_second": 0.007
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "Qwen/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": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.49.0.dev0",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.49.0.dev0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:638e8740f6cc71cc7fb8a19834065caed01ad6ab81a6c8d5d6bd25053cc2c9b3
+size 3554214752

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 sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
+size 11422063

tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "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",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 0.0,
+    "train_loss": 0.017702588886127465,
+    "train_runtime": 38308.166,
+    "train_samples": 72441,
+    "train_samples_per_second": 1.891,
+    "train_steps_per_second": 0.007
+}

trainer_state.json

@@ -0,0 +1,804 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 100,
+  "global_step": 283,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 387.2747495651245,
+      "epoch": 0.0176678445229682,
+      "grad_norm": 0.47285208106040955,
+      "kl": 0.0005104541778564453,
+      "learning_rate": 3.448275862068966e-06,
+      "loss": 0.0,
+      "reward": 0.6333333507180214,
+      "reward_std": 0.37065261919051407,
+      "rewards/accuracy_reward": 0.154427087912336,
+      "rewards/format_reward": 0.47890626452863216,
+      "step": 5
+    },
+    {
+      "completion_length": 235.940371799469,
+      "epoch": 0.0353356890459364,
+      "grad_norm": 0.3293702006340027,
+      "kl": 0.11603965759277343,
+      "learning_rate": 6.896551724137932e-06,
+      "loss": 0.0046,
+      "reward": 0.944270858168602,
+      "reward_std": 0.226960411388427,
+      "rewards/accuracy_reward": 0.08046875200234353,
+      "rewards/format_reward": 0.8638021022081375,
+      "step": 10
+    },
+    {
+      "completion_length": 178.96458854675294,
+      "epoch": 0.053003533568904596,
+      "grad_norm": 0.7117317318916321,
+      "kl": 0.06852264404296875,
+      "learning_rate": 1.0344827586206898e-05,
+      "loss": 0.0027,
+      "reward": 0.9953125268220901,
+      "reward_std": 0.19640953689813614,
+      "rewards/accuracy_reward": 0.07968750246800482,
+      "rewards/format_reward": 0.9156250216066837,
+      "step": 15
+    },
+    {
+      "completion_length": 97.98463859558106,
+      "epoch": 0.0706713780918728,
+      "grad_norm": 0.47843003273010254,
+      "kl": 0.14945068359375,
+      "learning_rate": 1.3793103448275863e-05,
+      "loss": 0.006,
+      "reward": 1.0330729484558105,
+      "reward_std": 0.19391254168003796,
+      "rewards/accuracy_reward": 0.10104166974779219,
+      "rewards/format_reward": 0.9320312693715096,
+      "step": 20
+    },
+    {
+      "completion_length": 156.7442744255066,
+      "epoch": 0.08833922261484099,
+      "grad_norm": 0.9008339047431946,
+      "kl": 0.190496826171875,
+      "learning_rate": 1.7241379310344828e-05,
+      "loss": 0.0076,
+      "reward": 1.098177120834589,
+      "reward_std": 0.19743250627070666,
+      "rewards/accuracy_reward": 0.13880208740010858,
+      "rewards/format_reward": 0.9593750193715096,
+      "step": 25
+    },
+    {
+      "completion_length": 168.8612027168274,
+      "epoch": 0.10600706713780919,
+      "grad_norm": 0.350289523601532,
+      "kl": 0.21292724609375,
+      "learning_rate": 1.999923511388017e-05,
+      "loss": 0.0085,
+      "reward": 1.0760417036712169,
+      "reward_std": 0.2866355457343161,
+      "rewards/accuracy_reward": 0.16901042133104055,
+      "rewards/format_reward": 0.9070312686264514,
+      "step": 30
+    },
+    {
+      "completion_length": 116.32890968322754,
+      "epoch": 0.12367491166077739,
+      "grad_norm": 0.80027836561203,
+      "kl": 0.30325927734375,
+      "learning_rate": 1.9972476383747748e-05,
+      "loss": 0.0121,
+      "reward": 1.1742187932133674,
+      "reward_std": 0.20648598652333022,
+      "rewards/accuracy_reward": 0.20156250540167092,
+      "rewards/format_reward": 0.9726562671363354,
+      "step": 35
+    },
+    {
+      "completion_length": 193.2869836807251,
+      "epoch": 0.1413427561837456,
+      "grad_norm": 0.6271685361862183,
+      "kl": 0.2177490234375,
+      "learning_rate": 1.9907590277344582e-05,
+      "loss": 0.0087,
+      "reward": 1.132031287252903,
+      "reward_std": 0.31216872576624155,
+      "rewards/accuracy_reward": 0.22682292172685264,
+      "rewards/format_reward": 0.9052083536982536,
+      "step": 40
+    },
+    {
+      "completion_length": 175.34974517822266,
+      "epoch": 0.15901060070671377,
+      "grad_norm": 6.180099964141846,
+      "kl": 0.3374267578125,
+      "learning_rate": 1.9804824871166254e-05,
+      "loss": 0.0135,
+      "reward": 1.1299479559063912,
+      "reward_std": 0.28119918145239353,
+      "rewards/accuracy_reward": 0.20546875577419996,
+      "rewards/format_reward": 0.9244791910052299,
+      "step": 45
+    },
+    {
+      "completion_length": 277.06511211395264,
+      "epoch": 0.17667844522968199,
+      "grad_norm": 0.4639797508716583,
+      "kl": 0.3301025390625,
+      "learning_rate": 1.9664573064143604e-05,
+      "loss": 0.0132,
+      "reward": 1.1270833693444728,
+      "reward_std": 0.327519618999213,
+      "rewards/accuracy_reward": 0.2442708406597376,
+      "rewards/format_reward": 0.8828125208616256,
+      "step": 50
+    },
+    {
+      "completion_length": 343.47344799041747,
+      "epoch": 0.19434628975265017,
+      "grad_norm": 0.21383723616600037,
+      "kl": 0.28677978515625,
+      "learning_rate": 1.948737107548771e-05,
+      "loss": 0.0115,
+      "reward": 1.1455729462206363,
+      "reward_std": 0.3933325769379735,
+      "rewards/accuracy_reward": 0.31302084140479564,
+      "rewards/format_reward": 0.8325521059334278,
+      "step": 55
+    },
+    {
+      "completion_length": 302.40521774291994,
+      "epoch": 0.21201413427561838,
+      "grad_norm": 0.38075944781303406,
+      "kl": 0.34866943359375,
+      "learning_rate": 1.9273896394584103e-05,
+      "loss": 0.0139,
+      "reward": 1.1281250447034836,
+      "reward_std": 0.40837063267827034,
+      "rewards/accuracy_reward": 0.29921875819563865,
+      "rewards/format_reward": 0.8289062693715096,
+      "step": 60
+    },
+    {
+      "completion_length": 315.8755304336548,
+      "epoch": 0.22968197879858657,
+      "grad_norm": 6.6895365715026855,
+      "kl": 0.87674560546875,
+      "learning_rate": 1.9024965190774262e-05,
+      "loss": 0.0351,
+      "reward": 1.225260455161333,
+      "reward_std": 0.32710962910205127,
+      "rewards/accuracy_reward": 0.3304687585681677,
+      "rewards/format_reward": 0.8947916865348816,
+      "step": 65
+    },
+    {
+      "completion_length": 247.97943458557128,
+      "epoch": 0.24734982332155478,
+      "grad_norm": 0.9354209303855896,
+      "kl": 0.59410400390625,
+      "learning_rate": 1.8741529192927528e-05,
+      "loss": 0.0238,
+      "reward": 1.1536458678543569,
+      "reward_std": 0.31826548781245945,
+      "rewards/accuracy_reward": 0.25130209028720857,
+      "rewards/format_reward": 0.9023437716066838,
+      "step": 70
+    },
+    {
+      "completion_length": 195.4330783843994,
+      "epoch": 0.26501766784452296,
+      "grad_norm": 1.8316482305526733,
+      "kl": 1.39520263671875,
+      "learning_rate": 1.8424672050733577e-05,
+      "loss": 0.0558,
+      "reward": 1.1125000342726707,
+      "reward_std": 0.29349851990118625,
+      "rewards/accuracy_reward": 0.206510423659347,
+      "rewards/format_reward": 0.9059896029531955,
+      "step": 75
+    },
+    {
+      "completion_length": 171.34349336624146,
+      "epoch": 0.2826855123674912,
+      "grad_norm": 1.258363127708435,
+      "kl": 4.63306884765625,
+      "learning_rate": 1.8075605191627242e-05,
+      "loss": 0.1854,
+      "reward": 1.1059896171092987,
+      "reward_std": 0.27957191290333866,
+      "rewards/accuracy_reward": 0.18489583707414567,
+      "rewards/format_reward": 0.9210937716066837,
+      "step": 80
+    },
+    {
+      "completion_length": 192.1846405982971,
+      "epoch": 0.3003533568904594,
+      "grad_norm": 2.0940563678741455,
+      "kl": 1.885009765625,
+      "learning_rate": 1.7695663189185703e-05,
+      "loss": 0.0754,
+      "reward": 1.0598958693444729,
+      "reward_std": 0.30320371966809034,
+      "rewards/accuracy_reward": 0.18020833837799727,
+      "rewards/format_reward": 0.8796875186264514,
+      "step": 85
+    },
+    {
+      "completion_length": 162.88151473999022,
+      "epoch": 0.31802120141342755,
+      "grad_norm": 0.3378926217556,
+      "kl": 0.84951171875,
+      "learning_rate": 1.7286298660705877e-05,
+      "loss": 0.034,
+      "reward": 1.1382812932133675,
+      "reward_std": 0.2608930370770395,
+      "rewards/accuracy_reward": 0.2187500058207661,
+      "rewards/format_reward": 0.9195312716066837,
+      "step": 90
+    },
+    {
+      "completion_length": 148.72239952087403,
+      "epoch": 0.33568904593639576,
+      "grad_norm": 0.2721768617630005,
+      "kl": 0.40316162109375,
+      "learning_rate": 1.6849076713469914e-05,
+      "loss": 0.0161,
+      "reward": 1.1757812909781933,
+      "reward_std": 0.23107599578797816,
+      "rewards/accuracy_reward": 0.21718750454019756,
+      "rewards/format_reward": 0.9585937716066837,
+      "step": 95
+    },
+    {
+      "completion_length": 177.43151569366455,
+      "epoch": 0.35335689045936397,
+      "grad_norm": 0.2728337049484253,
+      "kl": 0.4178466796875,
+      "learning_rate": 1.6385668960932143e-05,
+      "loss": 0.0167,
+      "reward": 1.1643229559063912,
+      "reward_std": 0.2606923753395677,
+      "rewards/accuracy_reward": 0.22552084056660532,
+      "rewards/format_reward": 0.9388021044433117,
+      "step": 100
+    },
+    {
+      "epoch": 0.35335689045936397,
+      "eval_completion_length": 138.00000286102295,
+      "eval_kl": 0.275390625,
+      "eval_loss": 0.010524081066250801,
+      "eval_reward": 1.2083333730697632,
+      "eval_reward_std": 0.22641220688819885,
+      "eval_rewards/accuracy_reward": 0.2500000074505806,
+      "eval_rewards/format_reward": 0.958333358168602,
+      "eval_runtime": 34.6506,
+      "eval_samples_per_second": 2.857,
+      "eval_steps_per_second": 0.058,
+      "step": 100
+    },
+    {
+      "completion_length": 178.4606819152832,
+      "epoch": 0.3710247349823322,
+      "grad_norm": 0.25652551651000977,
+      "kl": 0.298876953125,
+      "learning_rate": 1.5897847131705194e-05,
+      "loss": 0.012,
+      "reward": 1.1841146260499955,
+      "reward_std": 0.25834862645715473,
+      "rewards/accuracy_reward": 0.2442708403803408,
+      "rewards/format_reward": 0.9398437693715096,
+      "step": 105
+    },
+    {
+      "completion_length": 175.13984909057618,
+      "epoch": 0.38869257950530034,
+      "grad_norm": 0.2811754643917084,
+      "kl": 0.305810546875,
+      "learning_rate": 1.5387476295779737e-05,
+      "loss": 0.0122,
+      "reward": 1.1822917066514491,
+      "reward_std": 0.2543776072561741,
+      "rewards/accuracy_reward": 0.2390625063329935,
+      "rewards/format_reward": 0.9432291850447655,
+      "step": 110
+    },
+    {
+      "completion_length": 150.67943181991578,
+      "epoch": 0.40636042402826855,
+      "grad_norm": 0.36152493953704834,
+      "kl": 0.329296875,
+      "learning_rate": 1.4856507733875837e-05,
+      "loss": 0.0132,
+      "reward": 1.203125037252903,
+      "reward_std": 0.2355541598983109,
+      "rewards/accuracy_reward": 0.25260417382232847,
+      "rewards/format_reward": 0.9505208507180214,
+      "step": 115
+    },
+    {
+      "completion_length": 183.14740161895753,
+      "epoch": 0.42402826855123676,
+      "grad_norm": 0.2622193694114685,
+      "kl": 0.31475830078125,
+      "learning_rate": 1.4306971477188223e-05,
+      "loss": 0.0126,
+      "reward": 1.167187536507845,
+      "reward_std": 0.2702443292364478,
+      "rewards/accuracy_reward": 0.24375000614672898,
+      "rewards/format_reward": 0.9234375186264515,
+      "step": 120
+    },
+    {
+      "completion_length": 222.23958892822264,
+      "epoch": 0.4416961130742049,
+      "grad_norm": 0.31279441714286804,
+      "kl": 0.3115966796875,
+      "learning_rate": 1.3740968546047935e-05,
+      "loss": 0.0125,
+      "reward": 1.1656250365078449,
+      "reward_std": 0.31282911766320465,
+      "rewards/accuracy_reward": 0.25703125782310965,
+      "rewards/format_reward": 0.9085937686264515,
+      "step": 125
+    },
+    {
+      "completion_length": 216.9421937942505,
+      "epoch": 0.45936395759717313,
+      "grad_norm": 0.277396023273468,
+      "kl": 1.419482421875,
+      "learning_rate": 1.3160662917174045e-05,
+      "loss": 0.057,
+      "reward": 1.2429687909781932,
+      "reward_std": 0.2754800360649824,
+      "rewards/accuracy_reward": 0.3109375089406967,
+      "rewards/format_reward": 0.9320312678813935,
+      "step": 130
+    },
+    {
+      "completion_length": 187.78490142822267,
+      "epoch": 0.47703180212014135,
+      "grad_norm": 0.24579375982284546,
+      "kl": 0.23065185546875,
+      "learning_rate": 1.2568273250226681e-05,
+      "loss": 0.0092,
+      "reward": 1.2528646260499954,
+      "reward_std": 0.22978760292753578,
+      "rewards/accuracy_reward": 0.2968750076368451,
+      "rewards/format_reward": 0.9559896036982536,
+      "step": 135
+    },
+    {
+      "completion_length": 186.7362024307251,
+      "epoch": 0.49469964664310956,
+      "grad_norm": 0.23179057240486145,
+      "kl": 0.256982421875,
+      "learning_rate": 1.1966064405292887e-05,
+      "loss": 0.0103,
+      "reward": 1.2002604588866235,
+      "reward_std": 0.25049506183713677,
+      "rewards/accuracy_reward": 0.25625000689178706,
+      "rewards/format_reward": 0.9440104387700557,
+      "step": 140
+    },
+    {
+      "completion_length": 219.64427661895752,
+      "epoch": 0.5123674911660777,
+      "grad_norm": 0.3110530376434326,
+      "kl": 0.28455810546875,
+      "learning_rate": 1.1356338783736256e-05,
+      "loss": 0.0114,
+      "reward": 1.1953125342726707,
+      "reward_std": 0.2978471522219479,
+      "rewards/accuracy_reward": 0.2716145904734731,
+      "rewards/format_reward": 0.9236979357898235,
+      "step": 145
+    },
+    {
+      "completion_length": 228.7257875442505,
+      "epoch": 0.5300353356890459,
+      "grad_norm": 1.3053722381591797,
+      "kl": 0.31739501953125,
+      "learning_rate": 1.0741427525516463e-05,
+      "loss": 0.0127,
+      "reward": 1.2382812932133676,
+      "reward_std": 0.28989008301869035,
+      "rewards/accuracy_reward": 0.3020833428949118,
+      "rewards/format_reward": 0.9361979328095913,
+      "step": 150
+    },
+    {
+      "completion_length": 223.18255844116212,
+      "epoch": 0.5477031802120141,
+      "grad_norm": 0.29895108938217163,
+      "kl": 0.3016845703125,
+      "learning_rate": 1.012368159663363e-05,
+      "loss": 0.0121,
+      "reward": 1.262239620089531,
+      "reward_std": 0.2619246819987893,
+      "rewards/accuracy_reward": 0.3143229253590107,
+      "rewards/format_reward": 0.9479166887700557,
+      "step": 155
+    },
+    {
+      "completion_length": 237.11250705718993,
+      "epoch": 0.5653710247349824,
+      "grad_norm": 0.20286865532398224,
+      "kl": 0.22286376953125,
+      "learning_rate": 9.505462800772612e-06,
+      "loss": 0.0089,
+      "reward": 1.2424479641020298,
+      "reward_std": 0.2639943272806704,
+      "rewards/accuracy_reward": 0.29869792638346554,
+      "rewards/format_reward": 0.9437500186264515,
+      "step": 160
+    },
+    {
+      "completion_length": 245.8158924102783,
+      "epoch": 0.5830388692579506,
+      "grad_norm": 0.22661983966827393,
+      "kl": 0.3927978515625,
+      "learning_rate": 8.889134749511956e-06,
+      "loss": 0.0157,
+      "reward": 1.204166702926159,
+      "reward_std": 0.29652637206017973,
+      "rewards/accuracy_reward": 0.281250006519258,
+      "rewards/format_reward": 0.9229166865348816,
+      "step": 165
+    },
+    {
+      "completion_length": 211.2911516189575,
+      "epoch": 0.6007067137809188,
+      "grad_norm": 0.23035921156406403,
+      "kl": 0.2618896484375,
+      "learning_rate": 8.277053825620836e-06,
+      "loss": 0.0105,
+      "reward": 1.1955729581415653,
+      "reward_std": 0.28516152277588847,
+      "rewards/accuracy_reward": 0.26875000772997737,
+      "rewards/format_reward": 0.9268229320645333,
+      "step": 170
+    },
+    {
+      "completion_length": 195.80260944366455,
+      "epoch": 0.6183745583038869,
+      "grad_norm": 0.19353926181793213,
+      "kl": 0.26929931640625,
+      "learning_rate": 7.671560173993588e-06,
+      "loss": 0.0108,
+      "reward": 1.229947952926159,
+      "reward_std": 0.2544336979277432,
+      "rewards/accuracy_reward": 0.28854167480021714,
+      "rewards/format_reward": 0.9414062708616256,
+      "step": 175
+    },
+    {
+      "completion_length": 198.75104637145995,
+      "epoch": 0.6360424028268551,
+      "grad_norm": 0.2160399854183197,
+      "kl": 0.2649169921875,
+      "learning_rate": 7.07496875466589e-06,
+      "loss": 0.0106,
+      "reward": 1.2463542066514492,
+      "reward_std": 0.258614054415375,
+      "rewards/accuracy_reward": 0.3005208419635892,
+      "rewards/format_reward": 0.9458333536982536,
+      "step": 180
+    },
+    {
+      "completion_length": 232.85729770660402,
+      "epoch": 0.6537102473498233,
+      "grad_norm": 0.354937344789505,
+      "kl": 0.26788330078125,
+      "learning_rate": 6.489560492119225e-06,
+      "loss": 0.0107,
+      "reward": 1.2122396260499955,
+      "reward_std": 0.3070452008396387,
+      "rewards/accuracy_reward": 0.3000000074505806,
+      "rewards/format_reward": 0.9122396044433116,
+      "step": 185
+    },
+    {
+      "completion_length": 232.70495376586913,
+      "epoch": 0.6713780918727915,
+      "grad_norm": 0.2672271430492401,
+      "kl": 0.24927978515625,
+      "learning_rate": 5.9175735547120975e-06,
+      "loss": 0.01,
+      "reward": 1.2088542051613331,
+      "reward_std": 0.3101141072809696,
+      "rewards/accuracy_reward": 0.29739584028720856,
+      "rewards/format_reward": 0.9114583522081375,
+      "step": 190
+    },
+    {
+      "completion_length": 215.45365238189697,
+      "epoch": 0.6890459363957597,
+      "grad_norm": 0.2216312438249588,
+      "kl": 0.230712890625,
+      "learning_rate": 5.361194797579108e-06,
+      "loss": 0.0092,
+      "reward": 1.2455729581415653,
+      "reward_std": 0.2722341738641262,
+      "rewards/accuracy_reward": 0.3078125095693395,
+      "rewards/format_reward": 0.9377604357898235,
+      "step": 195
+    },
+    {
+      "completion_length": 229.8802146911621,
+      "epoch": 0.7067137809187279,
+      "grad_norm": 0.21737241744995117,
+      "kl": 0.29151611328125,
+      "learning_rate": 4.8225514017138205e-06,
+      "loss": 0.0117,
+      "reward": 1.2289062894880771,
+      "reward_std": 0.2966148443520069,
+      "rewards/accuracy_reward": 0.30390625931322574,
+      "rewards/format_reward": 0.9250000208616257,
+      "step": 200
+    },
+    {
+      "epoch": 0.7067137809187279,
+      "eval_completion_length": 219.22396087646484,
+      "eval_kl": 0.28564453125,
+      "eval_loss": 0.010753071866929531,
+      "eval_reward": 1.1796875298023224,
+      "eval_reward_std": 0.3818642348051071,
+      "eval_rewards/accuracy_reward": 0.2994791753590107,
+      "eval_rewards/format_reward": 0.880208358168602,
+      "eval_runtime": 37.6601,
+      "eval_samples_per_second": 2.629,
+      "eval_steps_per_second": 0.053,
+      "step": 200
+    },
+    {
+      "completion_length": 226.20911989212036,
+      "epoch": 0.7243816254416962,
+      "grad_norm": 0.32156360149383545,
+      "kl": 0.2999755859375,
+      "learning_rate": 4.303702741201431e-06,
+      "loss": 0.012,
+      "reward": 1.1695312850177288,
+      "reward_std": 0.3435072047635913,
+      "rewards/accuracy_reward": 0.2841145917773247,
+      "rewards/format_reward": 0.8854166857898236,
+      "step": 205
+    },
+    {
+      "completion_length": 182.92682847976684,
+      "epoch": 0.7420494699646644,
+      "grad_norm": 0.23930855095386505,
+      "kl": 0.3144775390625,
+      "learning_rate": 3.8066325096949153e-06,
+      "loss": 0.0126,
+      "reward": 1.2223958767950536,
+      "reward_std": 0.2844924574717879,
+      "rewards/accuracy_reward": 0.30312500772997736,
+      "rewards/format_reward": 0.9192708544433117,
+      "step": 210
+    },
+    {
+      "completion_length": 195.9677137374878,
+      "epoch": 0.7597173144876325,
+      "grad_norm": 0.27311936020851135,
+      "kl": 0.3186767578125,
+      "learning_rate": 3.3332411362372063e-06,
+      "loss": 0.0128,
+      "reward": 1.2109375447034836,
+      "reward_std": 0.2970659025013447,
+      "rewards/accuracy_reward": 0.29010417480021716,
+      "rewards/format_reward": 0.9208333544433117,
+      "step": 215
+    },
+    {
+      "completion_length": 229.93203754425048,
+      "epoch": 0.7773851590106007,
+      "grad_norm": 0.2597825825214386,
+      "kl": 0.2974365234375,
+      "learning_rate": 2.8853385194256677e-06,
+      "loss": 0.0119,
+      "reward": 1.1947917096316814,
+      "reward_std": 0.30680348305031657,
+      "rewards/accuracy_reward": 0.2929687576368451,
+      "rewards/format_reward": 0.9018229387700558,
+      "step": 220
+    },
+    {
+      "completion_length": 228.0653715133667,
+      "epoch": 0.7950530035335689,
+      "grad_norm": 0.29972967505455017,
+      "kl": 0.315771484375,
+      "learning_rate": 2.464637107698046e-06,
+      "loss": 0.0126,
+      "reward": 1.2190104603767395,
+      "reward_std": 0.2974198036827147,
+      "rewards/accuracy_reward": 0.30390625945292415,
+      "rewards/format_reward": 0.9151041880249977,
+      "step": 225
+    },
+    {
+      "completion_length": 203.7781304359436,
+      "epoch": 0.8127208480565371,
+      "grad_norm": 0.2677193880081177,
+      "kl": 0.29644775390625,
+      "learning_rate": 2.072745352195794e-06,
+      "loss": 0.0119,
+      "reward": 1.244531287997961,
+      "reward_std": 0.26717541785910726,
+      "rewards/accuracy_reward": 0.31093750847503543,
+      "rewards/format_reward": 0.9335937686264515,
+      "step": 230
+    },
+    {
+      "completion_length": 210.87682857513428,
+      "epoch": 0.8303886925795053,
+      "grad_norm": 0.22404737770557404,
+      "kl": 0.29898681640625,
+      "learning_rate": 1.7111615572361628e-06,
+      "loss": 0.012,
+      "reward": 1.2434896305203438,
+      "reward_std": 0.2834694870747626,
+      "rewards/accuracy_reward": 0.3164062585681677,
+      "rewards/format_reward": 0.9270833536982537,
+      "step": 235
+    },
+    {
+      "completion_length": 207.18307876586914,
+      "epoch": 0.8480565371024735,
+      "grad_norm": 0.24194031953811646,
+      "kl": 0.29622802734375,
+      "learning_rate": 1.381268151904298e-06,
+      "loss": 0.0119,
+      "reward": 1.2510417073965072,
+      "reward_std": 0.28840738208964467,
+      "rewards/accuracy_reward": 0.32239584159106016,
+      "rewards/format_reward": 0.9286458559334279,
+      "step": 240
+    },
+    {
+      "completion_length": 242.6622465133667,
+      "epoch": 0.8657243816254417,
+      "grad_norm": 0.3287613093852997,
+      "kl": 0.2822021484375,
+      "learning_rate": 1.0843264046665558e-06,
+      "loss": 0.0113,
+      "reward": 1.2122396215796472,
+      "reward_std": 0.3167353693395853,
+      "rewards/accuracy_reward": 0.3070312585681677,
+      "rewards/format_reward": 0.9052083536982536,
+      "step": 245
+    },
+    {
+      "completion_length": 238.33620433807374,
+      "epoch": 0.8833922261484098,
+      "grad_norm": 0.2897244691848755,
+      "kl": 0.2667236328125,
+      "learning_rate": 8.214716012124491e-07,
+      "loss": 0.0107,
+      "reward": 1.2494792073965073,
+      "reward_std": 0.2997074660845101,
+      "rewards/accuracy_reward": 0.3257812583819032,
+      "rewards/format_reward": 0.9236979350447655,
+      "step": 250
+    },
+    {
+      "completion_length": 235.88646545410157,
+      "epoch": 0.901060070671378,
+      "grad_norm": 0.23302209377288818,
+      "kl": 0.2700439453125,
+      "learning_rate": 5.937087039615619e-07,
+      "loss": 0.0108,
+      "reward": 1.2510417073965072,
+      "reward_std": 0.294480434525758,
+      "rewards/accuracy_reward": 0.32864584103226663,
+      "rewards/format_reward": 0.9223958544433117,
+      "step": 255
+    },
+    {
+      "completion_length": 236.86875438690186,
+      "epoch": 0.9187279151943463,
+      "grad_norm": 0.2569979429244995,
+      "kl": 0.2585693359375,
+      "learning_rate": 4.019085098303077e-07,
+      "loss": 0.0103,
+      "reward": 1.2484375432133674,
+      "reward_std": 0.28816566420719025,
+      "rewards/accuracy_reward": 0.3283854270353913,
+      "rewards/format_reward": 0.9200521036982536,
+      "step": 260
+    },
+    {
+      "completion_length": 247.26276741027831,
+      "epoch": 0.9363957597173145,
+      "grad_norm": 0.2708159387111664,
+      "kl": 0.26309814453125,
+      "learning_rate": 2.4680432094837394e-07,
+      "loss": 0.0105,
+      "reward": 1.2283854521811008,
+      "reward_std": 0.3015116843394935,
+      "rewards/accuracy_reward": 0.3195312600582838,
+      "rewards/format_reward": 0.9088541895151139,
+      "step": 265
+    },
+    {
+      "completion_length": 229.87032089233398,
+      "epoch": 0.9540636042402827,
+      "grad_norm": 0.4006361663341522,
+      "kl": 0.26412353515625,
+      "learning_rate": 1.289891410535593e-07,
+      "loss": 0.0106,
+      "reward": 1.2627604581415652,
+      "reward_std": 0.29111371459439395,
+      "rewards/accuracy_reward": 0.332812509033829,
+      "rewards/format_reward": 0.9299479357898235,
+      "step": 270
+    },
+    {
+      "completion_length": 230.29792442321778,
+      "epoch": 0.9717314487632509,
+      "grad_norm": 0.22953622043132782,
+      "kl": 0.26551513671875,
+      "learning_rate": 4.8913408283934874e-08,
+      "loss": 0.0106,
+      "reward": 1.2421875402331353,
+      "reward_std": 0.29439195804297924,
+      "rewards/accuracy_reward": 0.3216145934537053,
+      "rewards/format_reward": 0.9205729342997074,
+      "step": 275
+    },
+    {
+      "completion_length": 236.29141235351562,
+      "epoch": 0.9893992932862191,
+      "grad_norm": 0.38577350974082947,
+      "kl": 0.26748046875,
+      "learning_rate": 6.883273035447335e-09,
+      "loss": 0.0107,
+      "reward": 1.2638021275401115,
+      "reward_std": 0.31011410811915996,
+      "rewards/accuracy_reward": 0.34505209205672144,
+      "rewards/format_reward": 0.9187500178813934,
+      "step": 280
+    },
+    {
+      "completion_length": 239.0188110669454,
+      "epoch": 1.0,
+      "kl": 0.24015299479166666,
+      "reward": 1.251736156642437,
+      "reward_std": 0.2989154208141069,
+      "rewards/accuracy_reward": 0.32986112032085657,
+      "rewards/format_reward": 0.9218750211099783,
+      "step": 283,
+      "total_flos": 0.0,
+      "train_loss": 0.017702588886127465,
+      "train_runtime": 38308.166,
+      "train_samples_per_second": 1.891,
+      "train_steps_per_second": 0.007
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 283,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca3217afe04714f2e07e3d76865a2ef007f1ef493d74f0ada01a36c553c86982
+size 7544