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: RLVR-pmdata
+tags:
+- generated_from_trainer
+- trl
+- grpo
+licence: license
+---
+
+# Model Card for RLVR-pmdata
+
+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="RyanFan/RLVR-pmdata", 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/ryanzhang123-the-university-of-auckland/RLCR/runs/nzjcqerr) 
+
+
+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: 4.48.3
+- Pytorch: 2.5.1
+- Datasets: 4.0.0
+- Tokenizers: 0.21.4
+
+## 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
+}

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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.3",
+  "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": "4.48.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7c5b20d1810ce201680570cc498981959856d7778eec726b87f509577af1748
+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
+}

trainer_state.json

@@ -0,0 +1,1369 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9790660225442833,
+  "eval_steps": 50,
+  "global_step": 228,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.025,
+      "completions/max_length": 1310.4,
+      "completions/max_terminated_length": 1006.8,
+      "completions/mean_length": 270.43125,
+      "completions/mean_terminated_length": 237.94033203125,
+      "completions/min_length": 30.8,
+      "completions/min_terminated_length": 30.8,
+      "epoch": 0.0322061191626409,
+      "grad_norm": 0.035579483956098557,
+      "learning_rate": 1e-06,
+      "loss": -0.0011,
+      "num_tokens": 154629.0,
+      "reward": 0.415625,
+      "reward_std": 0.12983438968658448,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.83125,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 5
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.028125,
+      "completions/max_length": 1536.0,
+      "completions/max_terminated_length": 1284.8,
+      "completions/mean_length": 262.8453125,
+      "completions/mean_terminated_length": 225.93606567382812,
+      "completions/min_length": 27.6,
+      "completions/min_terminated_length": 27.6,
+      "epoch": 0.1288244766505636,
+      "grad_norm": 0.005246501415967941,
+      "learning_rate": 8.333333333333333e-07,
+      "loss": 0.0155,
+      "num_tokens": 800930.0,
+      "reward": 0.43046875,
+      "reward_std": 0.1471163362264633,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.8609375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 10
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.009375,
+      "completions/max_length": 1399.0,
+      "completions/max_terminated_length": 1281.0,
+      "completions/mean_length": 232.3828125,
+      "completions/mean_terminated_length": 219.98731079101563,
+      "completions/min_length": 19.6,
+      "completions/min_terminated_length": 19.6,
+      "epoch": 0.1932367149758454,
+      "grad_norm": 0.33856287598609924,
+      "learning_rate": 1e-06,
+      "loss": 0.0134,
+      "num_tokens": 1427423.0,
+      "reward": 0.4484375,
+      "reward_std": 0.11558105051517487,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.896875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 15
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.025,
+      "completions/max_length": 1536.0,
+      "completions/max_terminated_length": 1241.4,
+      "completions/mean_length": 230.415625,
+      "completions/mean_terminated_length": 196.781982421875,
+      "completions/min_length": 21.0,
+      "completions/min_terminated_length": 21.0,
+      "epoch": 0.2576489533011272,
+      "grad_norm": 0.00546436570584774,
+      "learning_rate": 1e-06,
+      "loss": 0.0153,
+      "num_tokens": 2065281.0,
+      "reward": 0.44921875,
+      "reward_std": 0.10368926376104355,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.8984375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 20
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0125,
+      "completions/max_length": 1536.0,
+      "completions/max_terminated_length": 1338.2,
+      "completions/mean_length": 229.59375,
+      "completions/mean_terminated_length": 213.0333465576172,
+      "completions/min_length": 19.8,
+      "completions/min_terminated_length": 19.8,
+      "epoch": 0.322061191626409,
+      "grad_norm": 0.0021914097014814615,
+      "learning_rate": 1e-06,
+      "loss": 0.0134,
+      "num_tokens": 2737021.0,
+      "reward": 0.465625,
+      "reward_std": 0.08104550689458848,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.93125,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 25
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0125,
+      "completions/max_length": 1536.0,
+      "completions/max_terminated_length": 1115.8,
+      "completions/mean_length": 204.8921875,
+      "completions/mean_terminated_length": 188.00881958007812,
+      "completions/min_length": 18.6,
+      "completions/min_terminated_length": 18.6,
+      "epoch": 0.3864734299516908,
+      "grad_norm": 0.31889012455940247,
+      "learning_rate": 1e-06,
+      "loss": 0.0083,
+      "num_tokens": 3360408.0,
+      "reward": 0.4734375,
+      "reward_std": 0.06902517676353455,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.946875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 30
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0109375,
+      "completions/max_length": 1464.2,
+      "completions/max_terminated_length": 1223.4,
+      "completions/mean_length": 178.346875,
+      "completions/mean_terminated_length": 163.26475830078124,
+      "completions/min_length": 19.0,
+      "completions/min_terminated_length": 19.0,
+      "epoch": 0.45088566827697263,
+      "grad_norm": 0.0017821220681071281,
+      "learning_rate": 1e-06,
+      "loss": 0.0029,
+      "num_tokens": 3953966.0,
+      "reward": 0.4875,
+      "reward_std": 0.03230288065969944,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.975,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 35
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0171875,
+      "completions/max_length": 1536.0,
+      "completions/max_terminated_length": 1127.8,
+      "completions/mean_length": 181.365625,
+      "completions/mean_terminated_length": 157.82559204101562,
+      "completions/min_length": 17.4,
+      "completions/min_terminated_length": 17.4,
+      "epoch": 0.5152979066022544,
+      "grad_norm": 0.0025463541969656944,
+      "learning_rate": 1e-06,
+      "loss": 0.0087,
+      "num_tokens": 4593824.0,
+      "reward": 0.48046875,
+      "reward_std": 0.05219025611877441,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9609375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 40
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0140625,
+      "completions/max_length": 1536.0,
+      "completions/max_terminated_length": 1005.2,
+      "completions/mean_length": 177.809375,
+      "completions/mean_terminated_length": 158.4675064086914,
+      "completions/min_length": 18.2,
+      "completions/min_terminated_length": 18.2,
+      "epoch": 0.5797101449275363,
+      "grad_norm": 0.0010303064482286572,
+      "learning_rate": 1e-06,
+      "loss": 0.0095,
+      "num_tokens": 5161622.0,
+      "reward": 0.48671875,
+      "reward_std": 0.037565046548843385,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9734375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 45
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0140625,
+      "completions/max_length": 1499.8,
+      "completions/max_terminated_length": 1231.2,
+      "completions/mean_length": 184.93125,
+      "completions/mean_terminated_length": 165.76856384277343,
+      "completions/min_length": 19.0,
+      "completions/min_terminated_length": 19.0,
+      "epoch": 0.644122383252818,
+      "grad_norm": 0.00440658675506711,
+      "learning_rate": 1e-06,
+      "loss": 0.0058,
+      "num_tokens": 5771802.0,
+      "reward": 0.4828125,
+      "reward_std": 0.04029847476631403,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.965625,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 50
+    },
+    {
+      "epoch": 0.644122383252818,
+      "eval_completions/clipped_ratio": 0.009963768115942028,
+      "eval_completions/max_length": 571.5652173913044,
+      "eval_completions/max_terminated_length": 503.2463768115942,
+      "eval_completions/mean_length": 159.33152173913044,
+      "eval_completions/mean_terminated_length": 147.455357675967,
+      "eval_completions/min_length": 39.65942028985507,
+      "eval_completions/min_terminated_length": 39.65942028985507,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 5771802.0,
+      "eval_reward": 0.4855072463768116,
+      "eval_reward_std": 0.03922215536020804,
+      "eval_rewards/accuracy_reward": 0.0,
+      "eval_rewards/brier_reward": 0.0,
+      "eval_rewards/confidence_one_or_zero": 0.0,
+      "eval_rewards/format_reward": 0.9710144927536232,
+      "eval_rewards/mean_confidence_reward": 0.0,
+      "eval_runtime": 458.7136,
+      "eval_samples_per_second": 0.301,
+      "eval_steps_per_second": 0.039,
+      "step": 50
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0109375,
+      "completions/max_length": 1460.2,
+      "completions/max_terminated_length": 1257.6,
+      "completions/mean_length": 168.40625,
+      "completions/mean_terminated_length": 153.32435913085936,
+      "completions/min_length": 18.4,
+      "completions/min_terminated_length": 18.4,
+      "epoch": 0.7085346215780999,
+      "grad_norm": 0.0,
+      "learning_rate": 1e-06,
+      "loss": 0.0042,
+      "num_tokens": 6394134.0,
+      "reward": 0.4875,
+      "reward_std": 0.03382910937070847,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.975,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 55
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0078125,
+      "completions/max_length": 1440.2,
+      "completions/max_terminated_length": 1214.0,
+      "completions/mean_length": 172.475,
+      "completions/mean_terminated_length": 162.49291381835937,
+      "completions/min_length": 17.4,
+      "completions/min_terminated_length": 17.4,
+      "epoch": 0.7729468599033816,
+      "grad_norm": 0.2661661207675934,
+      "learning_rate": 1e-06,
+      "loss": 0.0038,
+      "num_tokens": 6965774.0,
+      "reward": 0.4859375,
+      "reward_std": 0.03672229778021574,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.971875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 60
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0109375,
+      "completions/max_length": 1536.0,
+      "completions/max_terminated_length": 1079.2,
+      "completions/mean_length": 162.890625,
+      "completions/mean_terminated_length": 147.69927520751952,
+      "completions/min_length": 22.0,
+      "completions/min_terminated_length": 22.0,
+      "epoch": 0.8373590982286635,
+      "grad_norm": 0.0013718578265979886,
+      "learning_rate": 1e-06,
+      "loss": 0.0054,
+      "num_tokens": 7534152.0,
+      "reward": 0.48984375,
+      "reward_std": 0.027199982292950155,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9796875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 65
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0078125,
+      "completions/max_length": 1425.2,
+      "completions/max_terminated_length": 1097.4,
+      "completions/mean_length": 147.8484375,
+      "completions/mean_terminated_length": 136.90765686035155,
+      "completions/min_length": 21.8,
+      "completions/min_terminated_length": 21.8,
+      "epoch": 0.9017713365539453,
+      "grad_norm": 0.0,
+      "learning_rate": 1e-06,
+      "loss": 0.0032,
+      "num_tokens": 8069063.0,
+      "reward": 0.48984375,
+      "reward_std": 0.027199983410537244,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9796875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 70
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0140625,
+      "completions/max_length": 1483.4,
+      "completions/max_terminated_length": 1025.8,
+      "completions/mean_length": 146.9515625,
+      "completions/mean_terminated_length": 127.07637023925781,
+      "completions/min_length": 22.6,
+      "completions/min_terminated_length": 22.6,
+      "epoch": 0.966183574879227,
+      "grad_norm": 0.0007732895319350064,
+      "learning_rate": 1e-06,
+      "loss": 0.006,
+      "num_tokens": 8632176.0,
+      "reward": 0.490625,
+      "reward_std": 0.026516503281891347,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.98125,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 75
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0078125,
+      "completions/max_length": 1351.0,
+      "completions/max_terminated_length": 1082.4,
+      "completions/mean_length": 152.284375,
+      "completions/mean_terminated_length": 141.77883911132812,
+      "completions/min_length": 19.4,
+      "completions/min_terminated_length": 19.4,
+      "epoch": 1.0515297906602254,
+      "grad_norm": 0.0026292074471712112,
+      "learning_rate": 1e-06,
+      "loss": 0.0069,
+      "num_tokens": 9244190.0,
+      "reward": 0.4875,
+      "reward_std": 0.03382910899817944,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.975,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 80
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0125,
+      "completions/max_length": 1522.4,
+      "completions/max_terminated_length": 964.4,
+      "completions/mean_length": 137.1328125,
+      "completions/mean_terminated_length": 119.32202758789063,
+      "completions/min_length": 23.8,
+      "completions/min_terminated_length": 23.8,
+      "epoch": 1.1159420289855073,
+      "grad_norm": 0.0015776819782331586,
+      "learning_rate": 1e-06,
+      "loss": 0.0057,
+      "num_tokens": 9821539.0,
+      "reward": 0.48984375,
+      "reward_std": 0.027199982851743697,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9796875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 85
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0125,
+      "completions/max_length": 1487.4,
+      "completions/max_terminated_length": 1213.4,
+      "completions/mean_length": 157.5234375,
+      "completions/mean_terminated_length": 139.98736267089845,
+      "completions/min_length": 18.6,
+      "completions/min_terminated_length": 18.6,
+      "epoch": 1.180354267310789,
+      "grad_norm": 0.0026901797391474247,
+      "learning_rate": 1e-06,
+      "loss": 0.0038,
+      "num_tokens": 10433074.0,
+      "reward": 0.4921875,
+      "reward_std": 0.022097086161375047,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.984375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 90
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.003125,
+      "completions/max_length": 1289.4,
+      "completions/max_terminated_length": 1170.0,
+      "completions/mean_length": 145.9109375,
+      "completions/mean_terminated_length": 141.56537780761718,
+      "completions/min_length": 20.8,
+      "completions/min_terminated_length": 20.8,
+      "epoch": 1.2447665056360708,
+      "grad_norm": 0.0015041164588183165,
+      "learning_rate": 1e-06,
+      "loss": 0.0027,
+      "num_tokens": 11015073.0,
+      "reward": 0.49609375,
+      "reward_std": 0.011048543266952038,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9921875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 95
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0125,
+      "completions/max_length": 1430.8,
+      "completions/max_terminated_length": 1069.8,
+      "completions/mean_length": 147.3671875,
+      "completions/mean_terminated_length": 129.90637969970703,
+      "completions/min_length": 21.6,
+      "completions/min_terminated_length": 21.6,
+      "epoch": 1.3091787439613527,
+      "grad_norm": 0.0012177121825516224,
+      "learning_rate": 1e-06,
+      "loss": 0.0045,
+      "num_tokens": 11587028.0,
+      "reward": 0.48984375,
+      "reward_std": 0.02719998303800821,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9796875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 100
+    },
+    {
+      "epoch": 1.3091787439613527,
+      "eval_completions/clipped_ratio": 0.01177536231884058,
+      "eval_completions/max_length": 512.1449275362319,
+      "eval_completions/max_terminated_length": 432.1304347826087,
+      "eval_completions/mean_length": 143.50996376811594,
+      "eval_completions/mean_terminated_length": 128.20479691546896,
+      "eval_completions/min_length": 38.8768115942029,
+      "eval_completions/min_terminated_length": 38.8768115942029,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 11587028.0,
+      "eval_reward": 0.4913949275362319,
+      "eval_reward_std": 0.022569278552048447,
+      "eval_rewards/accuracy_reward": 0.0,
+      "eval_rewards/brier_reward": 0.0,
+      "eval_rewards/confidence_one_or_zero": 0.0,
+      "eval_rewards/format_reward": 0.9827898550724637,
+      "eval_rewards/mean_confidence_reward": 0.0,
+      "eval_runtime": 402.0293,
+      "eval_samples_per_second": 0.343,
+      "eval_steps_per_second": 0.045,
+      "step": 100
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.003125,
+      "completions/max_length": 1192.2,
+      "completions/max_terminated_length": 1007.8,
+      "completions/mean_length": 130.6890625,
+      "completions/mean_terminated_length": 126.32133483886719,
+      "completions/min_length": 23.8,
+      "completions/min_terminated_length": 23.8,
+      "epoch": 1.3735909822866343,
+      "grad_norm": 0.0019685616716742516,
+      "learning_rate": 1e-06,
+      "loss": 0.0014,
+      "num_tokens": 12151989.0,
+      "reward": 0.496875,
+      "reward_std": 0.00883883461356163,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.99375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 105
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0078125,
+      "completions/max_length": 1458.8,
+      "completions/max_terminated_length": 1068.6,
+      "completions/mean_length": 123.403125,
+      "completions/mean_terminated_length": 112.2623046875,
+      "completions/min_length": 21.4,
+      "completions/min_terminated_length": 21.4,
+      "epoch": 1.4380032206119162,
+      "grad_norm": 0.001040562056005001,
+      "learning_rate": 1e-06,
+      "loss": 0.0039,
+      "num_tokens": 12721343.0,
+      "reward": 0.49453125,
+      "reward_std": 0.015467960573732852,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9890625,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 110
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.009375,
+      "completions/max_length": 1273.0,
+      "completions/max_terminated_length": 1066.8,
+      "completions/mean_length": 138.396875,
+      "completions/mean_terminated_length": 125.22679138183594,
+      "completions/min_length": 24.0,
+      "completions/min_terminated_length": 24.0,
+      "epoch": 1.502415458937198,
+      "grad_norm": 0.002089962363243103,
+      "learning_rate": 1e-06,
+      "loss": 0.0033,
+      "num_tokens": 13313469.0,
+      "reward": 0.49296875,
+      "reward_std": 0.019887377880513667,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9859375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 115
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.003125,
+      "completions/max_length": 1064.0,
+      "completions/max_terminated_length": 1004.6,
+      "completions/mean_length": 136.971875,
+      "completions/mean_terminated_length": 133.62301940917968,
+      "completions/min_length": 23.2,
+      "completions/min_terminated_length": 23.2,
+      "epoch": 1.5668276972624797,
+      "grad_norm": 0.33233243227005005,
+      "learning_rate": 1e-06,
+      "loss": 0.0022,
+      "num_tokens": 13879963.0,
+      "reward": 0.49609375,
+      "reward_std": 0.011048543080687524,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9921875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 120
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.009375,
+      "completions/max_length": 1188.4,
+      "completions/max_terminated_length": 888.6,
+      "completions/mean_length": 123.18125,
+      "completions/mean_terminated_length": 111.09695281982422,
+      "completions/min_length": 20.6,
+      "completions/min_terminated_length": 20.6,
+      "epoch": 1.6312399355877618,
+      "grad_norm": 0.0011026322608813643,
+      "learning_rate": 1e-06,
+      "loss": 0.0032,
+      "num_tokens": 14404831.0,
+      "reward": 0.4953125,
+      "reward_std": 0.013258251920342445,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.990625,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 125
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0046875,
+      "completions/max_length": 1117.8,
+      "completions/max_terminated_length": 1037.2,
+      "completions/mean_length": 122.975,
+      "completions/mean_terminated_length": 116.31961364746094,
+      "completions/min_length": 17.6,
+      "completions/min_terminated_length": 17.6,
+      "epoch": 1.6956521739130435,
+      "grad_norm": 0.0014723028289154172,
+      "learning_rate": 1e-06,
+      "loss": 0.0027,
+      "num_tokens": 14976751.0,
+      "reward": 0.49609375,
+      "reward_std": 0.011048543080687524,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9921875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 130
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0046875,
+      "completions/max_length": 1286.2,
+      "completions/max_terminated_length": 984.2,
+      "completions/mean_length": 109.5984375,
+      "completions/mean_terminated_length": 102.81706085205079,
+      "completions/min_length": 20.2,
+      "completions/min_terminated_length": 20.2,
+      "epoch": 1.7600644122383253,
+      "grad_norm": 0.0021880464628338814,
+      "learning_rate": 1e-06,
+      "loss": 0.0011,
+      "num_tokens": 15543422.0,
+      "reward": 0.49609375,
+      "reward_std": 0.009522313810884952,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9921875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 135
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0046875,
+      "completions/max_length": 1336.0,
+      "completions/max_terminated_length": 1114.0,
+      "completions/mean_length": 128.7359375,
+      "completions/mean_terminated_length": 122.13045043945313,
+      "completions/min_length": 19.2,
+      "completions/min_terminated_length": 19.2,
+      "epoch": 1.8244766505636072,
+      "grad_norm": 0.000525438750628382,
+      "learning_rate": 1e-06,
+      "loss": 0.0026,
+      "num_tokens": 16118933.0,
+      "reward": 0.496875,
+      "reward_std": 0.00883883461356163,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.99375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 140
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.00625,
+      "completions/max_length": 1218.6,
+      "completions/max_terminated_length": 880.0,
+      "completions/mean_length": 127.0296875,
+      "completions/mean_terminated_length": 118.19228668212891,
+      "completions/min_length": 22.4,
+      "completions/min_terminated_length": 22.4,
+      "epoch": 1.8888888888888888,
+      "grad_norm": 0.0006711249006912112,
+      "learning_rate": 1e-06,
+      "loss": 0.0032,
+      "num_tokens": 16660720.0,
+      "reward": 0.4953125,
+      "reward_std": 0.01325825173407793,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.990625,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 145
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.00625,
+      "completions/max_length": 1304.8,
+      "completions/max_terminated_length": 1041.2,
+      "completions/mean_length": 131.8859375,
+      "completions/mean_terminated_length": 123.68300018310546,
+      "completions/min_length": 16.8,
+      "completions/min_terminated_length": 16.8,
+      "epoch": 1.9533011272141707,
+      "grad_norm": 0.4140954911708832,
+      "learning_rate": 1e-06,
+      "loss": 0.0023,
+      "num_tokens": 17209407.0,
+      "reward": 0.49609375,
+      "reward_std": 0.011048543266952038,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9921875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 150
+    },
+    {
+      "epoch": 1.9533011272141707,
+      "eval_completions/clipped_ratio": 0.004528985507246377,
+      "eval_completions/max_length": 394.6231884057971,
+      "eval_completions/max_terminated_length": 367.69565217391306,
+      "eval_completions/mean_length": 114.42391304347827,
+      "eval_completions/mean_terminated_length": 110.32013475722161,
+      "eval_completions/min_length": 34.32608695652174,
+      "eval_completions/min_terminated_length": 34.32608695652174,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 17209407.0,
+      "eval_reward": 0.49682971014492755,
+      "eval_reward_std": 0.008966933665932089,
+      "eval_rewards/accuracy_reward": 0.0,
+      "eval_rewards/brier_reward": 0.0,
+      "eval_rewards/confidence_one_or_zero": 0.0,
+      "eval_rewards/format_reward": 0.9936594202898551,
+      "eval_rewards/mean_confidence_reward": 0.0,
+      "eval_runtime": 318.6923,
+      "eval_samples_per_second": 0.433,
+      "eval_steps_per_second": 0.056,
+      "step": 150
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0015625,
+      "completions/max_length": 1172.2,
+      "completions/max_terminated_length": 959.6,
+      "completions/mean_length": 110.8133331298828,
+      "completions/mean_terminated_length": 108.54268798828124,
+      "completions/min_length": 20.0,
+      "completions/min_terminated_length": 20.0,
+      "epoch": 2.0386473429951693,
+      "grad_norm": 0.002583252266049385,
+      "learning_rate": 1e-06,
+      "loss": 0.0008,
+      "num_tokens": 17743744.0,
+      "reward": 0.4984375,
+      "reward_std": 0.004419417306780815,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.996875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 155
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0046875,
+      "completions/max_length": 1262.2,
+      "completions/max_terminated_length": 1153.0,
+      "completions/mean_length": 119.7703125,
+      "completions/mean_terminated_length": 113.08980102539063,
+      "completions/min_length": 20.0,
+      "completions/min_terminated_length": 20.0,
+      "epoch": 2.103059581320451,
+      "grad_norm": 0.5405756831169128,
+      "learning_rate": 1e-06,
+      "loss": 0.0021,
+      "num_tokens": 18318485.0,
+      "reward": 0.49765625,
+      "reward_std": 0.006629125773906707,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9953125,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 160
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0046875,
+      "completions/max_length": 1343.2,
+      "completions/max_terminated_length": 1049.6,
+      "completions/mean_length": 122.5515625,
+      "completions/mean_terminated_length": 115.89584350585938,
+      "completions/min_length": 19.8,
+      "completions/min_terminated_length": 19.8,
+      "epoch": 2.1674718196457325,
+      "grad_norm": 0.0,
+      "learning_rate": 1e-06,
+      "loss": 0.0024,
+      "num_tokens": 18834918.0,
+      "reward": 0.49609375,
+      "reward_std": 0.011048543266952038,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9921875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 165
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0015625,
+      "completions/max_length": 1011.8,
+      "completions/max_terminated_length": 898.2,
+      "completions/mean_length": 108.3484375,
+      "completions/mean_terminated_length": 106.08817596435547,
+      "completions/min_length": 17.0,
+      "completions/min_terminated_length": 17.0,
+      "epoch": 2.2318840579710146,
+      "grad_norm": 0.0007173310150392354,
+      "learning_rate": 1e-06,
+      "loss": 0.0006,
+      "num_tokens": 19381893.0,
+      "reward": 0.49921875,
+      "reward_std": 0.0022097086533904076,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9984375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 170
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.003125,
+      "completions/max_length": 1189.4,
+      "completions/max_terminated_length": 978.8,
+      "completions/mean_length": 119.48125,
+      "completions/mean_terminated_length": 115.07453308105468,
+      "completions/min_length": 18.4,
+      "completions/min_terminated_length": 18.4,
+      "epoch": 2.2962962962962963,
+      "grad_norm": 0.0,
+      "learning_rate": 1e-06,
+      "loss": 0.001,
+      "num_tokens": 19936521.0,
+      "reward": 0.49765625,
+      "reward_std": 0.0066291259601712225,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9953125,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 175
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.00625,
+      "completions/max_length": 1237.4,
+      "completions/max_terminated_length": 973.4,
+      "completions/mean_length": 106.9171875,
+      "completions/mean_terminated_length": 97.92234344482422,
+      "completions/min_length": 18.6,
+      "completions/min_terminated_length": 18.6,
+      "epoch": 2.360708534621578,
+      "grad_norm": 0.0012030096258968115,
+      "learning_rate": 1e-06,
+      "loss": 0.0034,
+      "num_tokens": 20497364.0,
+      "reward": 0.49296875,
+      "reward_std": 0.019887377507984638,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9859375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 180
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.00625,
+      "completions/max_length": 1361.2,
+      "completions/max_terminated_length": 1078.8,
+      "completions/mean_length": 117.1703125,
+      "completions/mean_terminated_length": 108.25378723144532,
+      "completions/min_length": 18.0,
+      "completions/min_terminated_length": 18.0,
+      "epoch": 2.42512077294686,
+      "grad_norm": 0.0008059889660216868,
+      "learning_rate": 1e-06,
+      "loss": 0.0036,
+      "num_tokens": 21029281.0,
+      "reward": 0.49453125,
+      "reward_std": 0.015467960573732852,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9890625,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 185
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0109375,
+      "completions/max_length": 1209.8,
+      "completions/max_terminated_length": 1021.0,
+      "completions/mean_length": 127.6078125,
+      "completions/mean_terminated_length": 112.13069000244141,
+      "completions/min_length": 16.8,
+      "completions/min_terminated_length": 16.8,
+      "epoch": 2.4895330112721417,
+      "grad_norm": 0.001932392013259232,
+      "learning_rate": 1e-06,
+      "loss": 0.0046,
+      "num_tokens": 21574886.0,
+      "reward": 0.49453125,
+      "reward_std": 0.015467960387468338,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9890625,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 190
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.00625,
+      "completions/max_length": 1176.8,
+      "completions/max_terminated_length": 840.4,
+      "completions/mean_length": 101.34375,
+      "completions/mean_terminated_length": 92.35088500976562,
+      "completions/min_length": 21.6,
+      "completions/min_terminated_length": 21.6,
+      "epoch": 2.5539452495974233,
+      "grad_norm": 0.0011432544561102986,
+      "learning_rate": 1e-06,
+      "loss": 0.0027,
+      "num_tokens": 22083498.0,
+      "reward": 0.496875,
+      "reward_std": 0.00883883461356163,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.99375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 195
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0109375,
+      "completions/max_length": 1407.4,
+      "completions/max_terminated_length": 902.4,
+      "completions/mean_length": 109.025,
+      "completions/mean_terminated_length": 93.25486755371094,
+      "completions/min_length": 16.8,
+      "completions/min_terminated_length": 16.8,
+      "epoch": 2.6183574879227054,
+      "grad_norm": 0.0014830685686320066,
+      "learning_rate": 1e-06,
+      "loss": 0.0037,
+      "num_tokens": 22645682.0,
+      "reward": 0.49375,
+      "reward_std": 0.01767766922712326,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 200
+    },
+    {
+      "epoch": 2.6183574879227054,
+      "eval_completions/clipped_ratio": 0.005434782608695652,
+      "eval_completions/max_length": 327.9130434782609,
+      "eval_completions/max_terminated_length": 296.4130434782609,
+      "eval_completions/mean_length": 100.78985507246377,
+      "eval_completions/mean_terminated_length": 95.00241558793662,
+      "eval_completions/min_length": 31.869565217391305,
+      "eval_completions/min_terminated_length": 31.869565217391305,
+      "eval_loss": 0.0,
+      "eval_num_tokens": 22645682.0,
+      "eval_reward": 0.4963768115942029,
+      "eval_reward_std": 0.009363153490467348,
+      "eval_rewards/accuracy_reward": 0.0,
+      "eval_rewards/brier_reward": 0.0,
+      "eval_rewards/confidence_one_or_zero": 0.0,
+      "eval_rewards/format_reward": 0.9927536231884058,
+      "eval_rewards/mean_confidence_reward": 0.0,
+      "eval_runtime": 264.3356,
+      "eval_samples_per_second": 0.522,
+      "eval_steps_per_second": 0.068,
+      "step": 200
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0015625,
+      "completions/max_length": 906.8,
+      "completions/max_terminated_length": 767.6,
+      "completions/mean_length": 94.16875,
+      "completions/mean_terminated_length": 91.92613220214844,
+      "completions/min_length": 17.6,
+      "completions/min_terminated_length": 17.6,
+      "epoch": 2.682769726247987,
+      "grad_norm": 0.0008456017822027206,
+      "learning_rate": 1e-06,
+      "loss": 0.001,
+      "num_tokens": 23170518.0,
+      "reward": 0.4984375,
+      "reward_std": 0.004419417306780815,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.996875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 205
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0015625,
+      "completions/max_length": 1083.0,
+      "completions/max_terminated_length": 987.2,
+      "completions/mean_length": 100.8828125,
+      "completions/mean_terminated_length": 98.64908905029297,
+      "completions/min_length": 17.4,
+      "completions/min_terminated_length": 17.4,
+      "epoch": 2.7471819645732687,
+      "grad_norm": 0.0,
+      "learning_rate": 1e-06,
+      "loss": 0.0018,
+      "num_tokens": 23710739.0,
+      "reward": 0.49609375,
+      "reward_std": 0.011048543080687524,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9921875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 210
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.003125,
+      "completions/max_length": 1102.0,
+      "completions/max_terminated_length": 789.8,
+      "completions/mean_length": 95.671875,
+      "completions/mean_terminated_length": 91.14502868652343,
+      "completions/min_length": 17.6,
+      "completions/min_terminated_length": 17.6,
+      "epoch": 2.8115942028985508,
+      "grad_norm": 0.0005856005591340363,
+      "learning_rate": 1e-06,
+      "loss": 0.0012,
+      "num_tokens": 24266489.0,
+      "reward": 0.4984375,
+      "reward_std": 0.004419417306780815,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.996875,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 215
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0125,
+      "completions/max_length": 1196.8,
+      "completions/max_terminated_length": 874.6,
+      "completions/mean_length": 112.078125,
+      "completions/mean_terminated_length": 95.37962341308594,
+      "completions/min_length": 16.2,
+      "completions/min_terminated_length": 16.2,
+      "epoch": 2.8760064412238324,
+      "grad_norm": 0.30407413840293884,
+      "learning_rate": 1e-06,
+      "loss": 0.0049,
+      "num_tokens": 24831699.0,
+      "reward": 0.4921875,
+      "reward_std": 0.022097086533904076,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.984375,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 220
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0046875,
+      "completions/max_length": 1315.4,
+      "completions/max_terminated_length": 829.2,
+      "completions/mean_length": 89.290625,
+      "completions/mean_terminated_length": 82.46375427246093,
+      "completions/min_length": 17.8,
+      "completions/min_terminated_length": 17.8,
+      "epoch": 2.940418679549114,
+      "grad_norm": 0.0008350868010893464,
+      "learning_rate": 1e-06,
+      "loss": 0.0016,
+      "num_tokens": 25352725.0,
+      "reward": 0.49765625,
+      "reward_std": 0.0066291259601712225,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9953125,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 225
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.005208333333333333,
+      "completions/max_length": 1218.0,
+      "completions/max_terminated_length": 1015.3333333333334,
+      "completions/mean_length": 110.06788380940755,
+      "completions/mean_terminated_length": 102.60350290934245,
+      "completions/min_length": 17.666666666666668,
+      "completions/min_terminated_length": 17.666666666666668,
+      "epoch": 2.9790660225442833,
+      "num_tokens": 25682129.0,
+      "reward": 0.4947916666666667,
+      "reward_std": 0.014731391022602717,
+      "rewards/accuracy_reward": 0.0,
+      "rewards/brier_reward": 0.0,
+      "rewards/confidence_one_or_zero": 0.0,
+      "rewards/format_reward": 0.9895833333333334,
+      "rewards/mean_confidence_reward": 0.0,
+      "step": 228,
+      "total_flos": 0.0,
+      "train_loss": 0.004439977674720515,
+      "train_runtime": 9596.7863,
+      "train_samples_per_second": 0.388,
+      "train_steps_per_second": 0.024
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 231,
+  "num_input_tokens_seen": 25682129,
+  "num_train_epochs": 3,
+  "save_steps": 60,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebfce4e612c23e2d7436653cfb2ec514c2a9b4250a4d15cd7a1d5c1d11f212fa
+size 7992