End of training

.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 +1,71 @@
+---
+base_model: Gensyn/Qwen2.5-0.5B-Instruct
+library_name: transformers
+model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-burrowing_timid_dog
+tags:
+- generated_from_trainer
+- rl-swarm
+- grpo
+- gensyn
+- I am burrowing timid dog
+- trl
+licence: license
+---
+
+# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-burrowing_timid_dog
+
+This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-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="DoomerHope/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-burrowing_timid_dog", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+ 
+
+
+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: 0.15.2
+- Transformers: 4.51.3
+- Pytorch: 2.5.1
+- Datasets: 3.5.0
+- Tokenizers: 0.21.1
+
+## 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": 0.06702521831626655,
+    "train_runtime": 3029.6015,
+    "train_samples": 115,
+    "train_samples_per_second": 1.056,
+    "train_steps_per_second": 0.066
+}

config.json

@@ -0,0 +1,28 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 896,
+  "initializer_range": 0.02,
+  "intermediate_size": 4864,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "qwen2",
+  "num_attention_heads": 14,
+  "num_hidden_layers": 24,
+  "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": "float32",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "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": "4.51.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3777fe2b2b304b2a7bcbb8a25ece35e073fc0ca20663d3f73caddad3a42e3aa
+size 1976163472

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

train_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 0.0,
+    "train_loss": 0.06702521831626655,
+    "train_runtime": 3029.6015,
+    "train_samples": 115,
+    "train_samples_per_second": 1.056,
+    "train_steps_per_second": 0.066
+}

trainer_state.json

@@ -0,0 +1,1943 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 13.347826086956522,
+  "eval_steps": 500,
+  "global_step": 200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 469.125,
+      "epoch": 0.1391304347826087,
+      "grad_norm": 6.651979446411133,
+      "kl": 0.0,
+      "learning_rate": 1.6666666666666665e-07,
+      "loss": -0.0,
+      "reward": 0.7131982836872339,
+      "reward_std": 0.6387130866860389,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.0625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.34851078875362873,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.23968750052154064,
+      "step": 2
+    },
+    {
+      "completion_length": 407.625,
+      "epoch": 0.2782608695652174,
+      "grad_norm": 9.905783653259277,
+      "kl": 0.005283124854031485,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": 1.395947689190507,
+      "reward_std": 1.2750358628109097,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.42385393101722,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.2845937702804804,
+      "step": 4
+    },
+    {
+      "completion_length": 322.78125,
+      "epoch": 0.41739130434782606,
+      "grad_norm": 6.311988353729248,
+      "kl": 0.0086102341338119,
+      "learning_rate": 8.333333333333333e-07,
+      "loss": 0.0,
+      "reward": 1.2708191433921456,
+      "reward_std": 0.6618030449608341,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.1875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.3736316340509802,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.4596874974668026,
+      "step": 6
+    },
+    {
+      "completion_length": 356.28125,
+      "epoch": 0.5565217391304348,
+      "grad_norm": 7.336427211761475,
+      "kl": 0.010165997344302014,
+      "learning_rate": 9.99934441832816e-07,
+      "loss": 0.0,
+      "reward": 1.0624355049803853,
+      "reward_std": 1.175302461721003,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.38565425004344434,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.2861562427133322,
+      "step": 8
+    },
+    {
+      "completion_length": 355.96875,
+      "epoch": 0.6956521739130435,
+      "grad_norm": 9.285393714904785,
+      "kl": 0.010603671733406372,
+      "learning_rate": 9.994100796397953e-07,
+      "loss": 0.0,
+      "reward": 1.2944127935916185,
+      "reward_std": 0.9169077111873776,
+      "rewards/concensus_correctness_reward_func": 0.06012500077486038,
+      "rewards/consensus_reward_func": 0.1875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.3885377976548625,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.5176250012591481,
+      "step": 10
+    },
+    {
+      "completion_length": 432.5,
+      "epoch": 0.8347826086956521,
+      "grad_norm": 6.065662384033203,
+      "kl": 0.022270067056524567,
+      "learning_rate": 9.983619052372847e-07,
+      "loss": 0.0,
+      "reward": 0.7962917103432119,
+      "reward_std": 0.9656887443270534,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.3581042473670095,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.06318750325590372,
+      "step": 12
+    },
+    {
+      "completion_length": 382.5,
+      "epoch": 0.9739130434782609,
+      "grad_norm": 8.355684280395508,
+      "kl": 0.021279162167047616,
+      "learning_rate": 9.967910180154888e-07,
+      "loss": 0.0,
+      "reward": 0.9942772081121802,
+      "reward_std": 1.3131475580157712,
+      "rewards/concensus_correctness_reward_func": 0.00787500012665987,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.3675272043328732,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.24387499503791332,
+      "step": 14
+    },
+    {
+      "completion_length": 485.40909090909093,
+      "epoch": 1.0695652173913044,
+      "grad_norm": 6.345098495483398,
+      "kl": 0.04893516427413984,
+      "learning_rate": 9.946990656181779e-07,
+      "loss": 0.0,
+      "reward": 1.2941602251746438,
+      "reward_std": 1.729960096153346,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.2727272727272727,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.2727272727272727,
+      "rewards/question_recreation_reward_func": 0.330114785073833,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.022727272727272728,
+      "rewards/xmlcount_reward_func": 0.3958636359734969,
+      "step": 16
+    },
+    {
+      "completion_length": 566.75,
+      "epoch": 1.208695652173913,
+      "grad_norm": 4.4084792137146,
+      "kl": 0.035054971085628495,
+      "learning_rate": 9.92088242214537e-07,
+      "loss": 0.0,
+      "reward": 0.9450036096386611,
+      "reward_std": 0.7893581000389531,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.43631610507145524,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.24306250270456076,
+      "step": 18
+    },
+    {
+      "completion_length": 460.875,
+      "epoch": 1.3478260869565217,
+      "grad_norm": 5.167897701263428,
+      "kl": 0.04613033159694169,
+      "learning_rate": 9.889612861977853e-07,
+      "loss": 0.0,
+      "reward": 0.4604216940351762,
+      "reward_std": 1.0246329813089687,
+      "rewards/concensus_correctness_reward_func": 0.07824999839067459,
+      "rewards/consensus_reward_func": 0.0,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.26920293824514374,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.11296873912215233,
+      "step": 20
+    },
+    {
+      "completion_length": 387.40625,
+      "epoch": 1.4869565217391305,
+      "grad_norm": 4.876871109008789,
+      "kl": 0.08161755069158971,
+      "learning_rate": 9.853214773129795e-07,
+      "loss": 0.0001,
+      "reward": 1.786725739017129,
+      "reward_std": 1.259733703918755,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.5388507556635886,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.6072499956935644,
+      "step": 22
+    },
+    {
+      "completion_length": 503.90625,
+      "epoch": 1.626086956521739,
+      "grad_norm": 11.22158145904541,
+      "kl": 0.08277971870847978,
+      "learning_rate": 9.81172633217015e-07,
+      "loss": 0.0001,
+      "reward": 1.2416393284220248,
+      "reward_std": 1.4355839109048247,
+      "rewards/concensus_correctness_reward_func": 0.00787500012665987,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.38088934967527166,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.0716249905526638,
+      "step": 24
+    },
+    {
+      "completion_length": 416.90625,
+      "epoch": 1.7652173913043478,
+      "grad_norm": 8.147880554199219,
+      "kl": 0.11381094309035689,
+      "learning_rate": 9.765191054744304e-07,
+      "loss": 0.0001,
+      "reward": 1.2789664026349783,
+      "reward_std": 1.0479871579445899,
+      "rewards/concensus_correctness_reward_func": 0.0078125,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5266226634848863,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.3695312385680154,
+      "step": 26
+    },
+    {
+      "completion_length": 442.40625,
+      "epoch": 1.9043478260869566,
+      "grad_norm": 9.084968566894531,
+      "kl": 0.12449931562878191,
+      "learning_rate": 9.713657749932171e-07,
+      "loss": 0.0001,
+      "reward": 1.339750369079411,
+      "reward_std": 1.2324395878240466,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.40315663220826536,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.49909375328570604,
+      "step": 28
+    },
+    {
+      "completion_length": 589.7272727272727,
+      "epoch": 2.0,
+      "grad_norm": 1.5676738023757935,
+      "kl": 0.0792797537181865,
+      "learning_rate": 9.657180469054212e-07,
+      "loss": 0.0001,
+      "reward": 0.922713640569286,
+      "reward_std": 1.0948863579383628,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.36363636363636365,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.22689545425501736,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.3321818167513067,
+      "step": 30
+    },
+    {
+      "completion_length": 454.40625,
+      "epoch": 2.139130434782609,
+      "grad_norm": 5.896516799926758,
+      "kl": 0.14648738969117403,
+      "learning_rate": 9.59581844897906e-07,
+      "loss": 0.0001,
+      "reward": 1.1079749837517738,
+      "reward_std": 1.25278989225626,
+      "rewards/concensus_correctness_reward_func": 0.023125000298023224,
+      "rewards/consensus_reward_func": 0.1875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.39031872944906354,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.22578124795109034,
+      "step": 32
+    },
+    {
+      "completion_length": 365.28125,
+      "epoch": 2.2782608695652176,
+      "grad_norm": 5.692052841186523,
+      "kl": 0.16005453513935208,
+      "learning_rate": 9.529636049992233e-07,
+      "loss": 0.0002,
+      "reward": 1.4690629467368126,
+      "reward_std": 1.0634890967048705,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.4147504305001348,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.5855625066906214,
+      "step": 34
+    },
+    {
+      "completion_length": 496.46875,
+      "epoch": 2.417391304347826,
+      "grad_norm": 3.690098762512207,
+      "kl": 0.116360651969444,
+      "learning_rate": 9.458702688291071e-07,
+      "loss": 0.0001,
+      "reward": 0.9128094336483628,
+      "reward_std": 1.1920369119616225,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.0625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.48627819842658937,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.1609062496572733,
+      "step": 36
+    },
+    {
+      "completion_length": 357.25,
+      "epoch": 2.5565217391304347,
+      "grad_norm": 5.994088649749756,
+      "kl": 0.22547357506118715,
+      "learning_rate": 9.383092763176738e-07,
+      "loss": 0.0002,
+      "reward": 1.4751352742314339,
+      "reward_std": 1.2913847951567732,
+      "rewards/concensus_correctness_reward_func": 0.12025000154972076,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.313854017527774,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.078125,
+      "rewards/xmlcount_reward_func": 0.5879062628373504,
+      "step": 38
+    },
+    {
+      "completion_length": 365.125,
+      "epoch": 2.6956521739130435,
+      "grad_norm": 5.768886566162109,
+      "kl": 0.20043913254630752,
+      "learning_rate": 9.302885579019626e-07,
+      "loss": 0.0002,
+      "reward": 1.5617408801335841,
+      "reward_std": 0.770021176605951,
+      "rewards/concensus_correctness_reward_func": 0.0078125,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.421834665350616,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0625,
+      "rewards/xmlcount_reward_func": 0.6320937536656857,
+      "step": 40
+    },
+    {
+      "completion_length": 439.46875,
+      "epoch": 2.8347826086956522,
+      "grad_norm": 6.0984392166137695,
+      "kl": 0.16734766645822674,
+      "learning_rate": 9.218165262080022e-07,
+      "loss": 0.0002,
+      "reward": 1.584086636081338,
+      "reward_std": 1.4578522881492972,
+      "rewards/concensus_correctness_reward_func": 0.03425000049173832,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5438053905963898,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.046875,
+      "rewards/xmlcount_reward_func": 0.5216562459245324,
+      "step": 42
+    },
+    {
+      "completion_length": 392.5,
+      "epoch": 2.973913043478261,
+      "grad_norm": 6.062408924102783,
+      "kl": 0.25865777861326933,
+      "learning_rate": 9.129020672271281e-07,
+      "loss": 0.0003,
+      "reward": 2.0376671738922596,
+      "reward_std": 0.9936516415327787,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.49360471079126,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.046875,
+      "rewards/xmlcount_reward_func": 0.809687502682209,
+      "step": 44
+    },
+    {
+      "completion_length": 411.45454545454544,
+      "epoch": 3.0695652173913044,
+      "grad_norm": 281.8279113769531,
+      "kl": 62.15620172836564,
+      "learning_rate": 9.035545309958046e-07,
+      "loss": 0.0427,
+      "reward": 1.3393509787592022,
+      "reward_std": 1.1218569095839153,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.2727272727272727,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.18181818181818182,
+      "rewards/question_recreation_reward_func": 0.42680550942366774,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09090909090909091,
+      "rewards/xmlcount_reward_func": 0.36709091744639655,
+      "step": 46
+    },
+    {
+      "completion_length": 447.8125,
+      "epoch": 3.208695652173913,
+      "grad_norm": 3.567924737930298,
+      "kl": 0.1984356112079695,
+      "learning_rate": 8.937837217887272e-07,
+      "loss": 0.0002,
+      "reward": 1.4719885361846536,
+      "reward_std": 1.0692140067085347,
+      "rewards/concensus_correctness_reward_func": 0.00787500012665987,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.4144260432221927,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0625,
+      "rewards/xmlcount_reward_func": 0.487187503837049,
+      "step": 48
+    },
+    {
+      "completion_length": 355.78125,
+      "epoch": 3.3478260869565215,
+      "grad_norm": 6.904684066772461,
+      "kl": 0.3303725360892713,
+      "learning_rate": 8.83599887835493e-07,
+      "loss": 0.0003,
+      "reward": 1.946224831044674,
+      "reward_std": 1.4707771465182304,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.4765685759484768,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.078125,
+      "rewards/xmlcount_reward_func": 0.7040312550961971,
+      "step": 50
+    },
+    {
+      "completion_length": 256.5,
+      "epoch": 3.4869565217391303,
+      "grad_norm": 5.664855003356934,
+      "kl": 0.36116583459079266,
+      "learning_rate": 8.73013710571623e-07,
+      "loss": 0.0004,
+      "reward": 1.7132504982873797,
+      "reward_std": 1.3052820832235739,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.1875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.5305317317834124,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.109375,
+      "rewards/xmlcount_reward_func": 0.698343763127923,
+      "step": 52
+    },
+    {
+      "completion_length": 416.03125,
+      "epoch": 3.626086956521739,
+      "grad_norm": 5.109403133392334,
+      "kl": 0.26479395059868693,
+      "learning_rate": 8.620362934352108e-07,
+      "loss": 0.0003,
+      "reward": 1.4901817552745342,
+      "reward_std": 1.71798007003963,
+      "rewards/concensus_correctness_reward_func": 0.12806250154972076,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.45561923924833536,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.37524999491870403,
+      "step": 54
+    },
+    {
+      "completion_length": 274.25,
+      "epoch": 3.765217391304348,
+      "grad_norm": 6.084560394287109,
+      "kl": 0.44902627915143967,
+      "learning_rate": 8.506791502209496e-07,
+      "loss": 0.0004,
+      "reward": 2.3243369851261377,
+      "reward_std": 1.0572451442712918,
+      "rewards/concensus_correctness_reward_func": 0.06306250067427754,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.47302445746026933,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 0.9601250104606152,
+      "step": 56
+    },
+    {
+      "completion_length": 282.21875,
+      "epoch": 3.9043478260869566,
+      "grad_norm": 5.800886631011963,
+      "kl": 0.3701688055880368,
+      "learning_rate": 8.389541930037516e-07,
+      "loss": 0.0004,
+      "reward": 2.083632607012987,
+      "reward_std": 0.8866405207663774,
+      "rewards/concensus_correctness_reward_func": 0.06793750077486038,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.5871013961732388,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 0.8504687547683716,
+      "step": 58
+    },
+    {
+      "completion_length": 188.04545454545453,
+      "epoch": 4.0,
+      "grad_norm": 3.667579174041748,
+      "kl": 0.5356881266290491,
+      "learning_rate": 8.268737196446263e-07,
+      "loss": 0.0004,
+      "reward": 2.0462357076731594,
+      "reward_std": 0.6923801770264452,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.36363636363636365,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.5550993525169112,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09090909090909091,
+      "rewards/xmlcount_reward_func": 1.0365909121253274,
+      "step": 60
+    },
+    {
+      "completion_length": 332.0,
+      "epoch": 4.139130434782609,
+      "grad_norm": 7.300347328186035,
+      "kl": 0.4795700521208346,
+      "learning_rate": 8.144504008919222e-07,
+      "loss": 0.0005,
+      "reward": 1.7712411060929298,
+      "reward_std": 0.9126198044978082,
+      "rewards/concensus_correctness_reward_func": 0.06012500077486038,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.6377411026624031,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.078125,
+      "rewards/xmlcount_reward_func": 0.6827500015497208,
+      "step": 62
+    },
+    {
+      "completion_length": 325.15625,
+      "epoch": 4.278260869565218,
+      "grad_norm": 5.623091697692871,
+      "kl": 0.46719347033649683,
+      "learning_rate": 8.016972670914623e-07,
+      "loss": 0.0005,
+      "reward": 2.370351441204548,
+      "reward_std": 0.9159036946948618,
+      "rewards/concensus_correctness_reward_func": 0.023125000298023224,
+      "rewards/consensus_reward_func": 0.5625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.5068202022230253,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 0.9341562539339066,
+      "step": 64
+    },
+    {
+      "completion_length": 237.28125,
+      "epoch": 4.417391304347826,
+      "grad_norm": 13.47927188873291,
+      "kl": 0.6326476326212287,
+      "learning_rate": 7.886276945195097e-07,
+      "loss": 0.0006,
+      "reward": 2.162361765280366,
+      "reward_std": 1.1361898751856643,
+      "rewards/concensus_correctness_reward_func": 0.04493749933317304,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.4671117801917717,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 0.9003124982118607,
+      "step": 66
+    },
+    {
+      "completion_length": 306.21875,
+      "epoch": 4.556521739130435,
+      "grad_norm": 6.569784641265869,
+      "kl": 0.6667315466329455,
+      "learning_rate": 7.752553913529018e-07,
+      "loss": 0.0007,
+      "reward": 2.39416229352355,
+      "reward_std": 0.8820177484303713,
+      "rewards/concensus_correctness_reward_func": 0.00787500012665987,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.5330685274675488,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 0.9782187566161156,
+      "step": 68
+    },
+    {
+      "completion_length": 252.125,
+      "epoch": 4.695652173913043,
+      "grad_norm": 9.948299407958984,
+      "kl": 0.5775284115225077,
+      "learning_rate": 7.61594383291065e-07,
+      "loss": 0.0006,
+      "reward": 1.8165949955582619,
+      "reward_std": 0.8422716949135065,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.4778762402711436,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 0.9637187533080578,
+      "step": 70
+    },
+    {
+      "completion_length": 305.875,
+      "epoch": 4.834782608695652,
+      "grad_norm": 6.913571357727051,
+      "kl": 0.5833932915702462,
+      "learning_rate": 7.476589988449938e-07,
+      "loss": 0.0006,
+      "reward": 1.5893209166824818,
+      "reward_std": 0.9271868723444641,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.1875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.46325845512910746,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.109375,
+      "rewards/xmlcount_reward_func": 0.7041874974966049,
+      "step": 72
+    },
+    {
+      "completion_length": 316.6875,
+      "epoch": 4.973913043478261,
+      "grad_norm": 8.020514488220215,
+      "kl": 1.4912343490868807,
+      "learning_rate": 7.334638543086203e-07,
+      "loss": 0.0015,
+      "reward": 2.227024871855974,
+      "reward_std": 1.0563541476149112,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.1875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.5625,
+      "rewards/question_recreation_reward_func": 0.4756498399656266,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.109375,
+      "rewards/xmlcount_reward_func": 0.8919999971985817,
+      "step": 74
+    },
+    {
+      "completion_length": 368.77272727272725,
+      "epoch": 5.069565217391304,
+      "grad_norm": 7.067770957946777,
+      "kl": 0.7251402681524103,
+      "learning_rate": 7.190238384283412e-07,
+      "loss": 0.0005,
+      "reward": 1.581421895460649,
+      "reward_std": 1.238969932902943,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.2727272727272727,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.3257855162532492,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.13636363636363635,
+      "rewards/xmlcount_reward_func": 0.8465454578399658,
+      "step": 76
+    },
+    {
+      "completion_length": 300.25,
+      "epoch": 5.208695652173913,
+      "grad_norm": 5.457042217254639,
+      "kl": 1.8599772108718753,
+      "learning_rate": 7.043540967867781e-07,
+      "loss": 0.0019,
+      "reward": 2.499551836401224,
+      "reward_std": 1.034086337662302,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.5,
+      "rewards/question_recreation_reward_func": 0.7101768124848604,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.8206249997019768,
+      "step": 78
+    },
+    {
+      "completion_length": 251.78125,
+      "epoch": 5.3478260869565215,
+      "grad_norm": 6.8296637535095215,
+      "kl": 1.4302785657346249,
+      "learning_rate": 6.894700159171534e-07,
+      "loss": 0.0014,
+      "reward": 2.0862809866666794,
+      "reward_std": 1.1299847126938403,
+      "rewards/concensus_correctness_reward_func": 0.0078125,
+      "rewards/consensus_reward_func": 0.5,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.45674972841516137,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.109375,
+      "rewards/xmlcount_reward_func": 1.012343741953373,
+      "step": 80
+    },
+    {
+      "completion_length": 277.28125,
+      "epoch": 5.48695652173913,
+      "grad_norm": 14.91946792602539,
+      "kl": 4.983186790719628,
+      "learning_rate": 6.743872071649411e-07,
+      "loss": 0.005,
+      "reward": 2.2909165136516094,
+      "reward_std": 0.771335305791581,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.5625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.5036664966028184,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.109375,
+      "rewards/xmlcount_reward_func": 0.8653750023804605,
+      "step": 82
+    },
+    {
+      "completion_length": 226.3125,
+      "epoch": 5.626086956521739,
+      "grad_norm": 6.772074222564697,
+      "kl": 0.7162865996360779,
+      "learning_rate": 6.59121490313722e-07,
+      "loss": 0.0007,
+      "reward": 2.433136470615864,
+      "reward_std": 1.169273011459154,
+      "rewards/concensus_correctness_reward_func": 0.023125000298023224,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.6323864688165486,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.15625,
+      "rewards/xmlcount_reward_func": 0.9963750019669533,
+      "step": 84
+    },
+    {
+      "completion_length": 312.71875,
+      "epoch": 5.765217391304348,
+      "grad_norm": 8.089889526367188,
+      "kl": 0.8305776445195079,
+      "learning_rate": 6.436888769924141e-07,
+      "loss": 0.0008,
+      "reward": 2.09120103251189,
+      "reward_std": 0.8542511662235484,
+      "rewards/concensus_correctness_reward_func": 0.002937499899417162,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5309822625713423,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 0.9166562501341105,
+      "step": 86
+    },
+    {
+      "completion_length": 222.25,
+      "epoch": 5.904347826086957,
+      "grad_norm": 8.415414810180664,
+      "kl": 0.8001231420785189,
+      "learning_rate": 6.281055538812861e-07,
+      "loss": 0.0008,
+      "reward": 3.005630634725094,
+      "reward_std": 1.0765592579264194,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.5625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.5625,
+      "rewards/question_recreation_reward_func": 0.7028806544840336,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 1.0527499988675117,
+      "step": 88
+    },
+    {
+      "completion_length": 244.8181818181818,
+      "epoch": 6.0,
+      "grad_norm": 4.685628890991211,
+      "kl": 1.208802805705504,
+      "learning_rate": 6.123878657343647e-07,
+      "loss": 0.0008,
+      "reward": 2.176835905421864,
+      "reward_std": 0.7615151606150903,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.45454545454545453,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.6479268099435351,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.06818181818181818,
+      "rewards/xmlcount_reward_func": 1.0061818198724226,
+      "step": 90
+    },
+    {
+      "completion_length": 267.21875,
+      "epoch": 6.139130434782609,
+      "grad_norm": 8.282288551330566,
+      "kl": 0.8628208544105291,
+      "learning_rate": 5.96552298236044e-07,
+      "loss": 0.0009,
+      "reward": 2.53002592921257,
+      "reward_std": 1.104838589206338,
+      "rewards/concensus_correctness_reward_func": 0.08362499997019768,
+      "rewards/consensus_reward_func": 0.5625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.5674946635845117,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 1.00390625,
+      "step": 92
+    },
+    {
+      "completion_length": 181.5,
+      "epoch": 6.278260869565218,
+      "grad_norm": 9.85130786895752,
+      "kl": 1.7928204946219921,
+      "learning_rate": 5.806154607098799e-07,
+      "loss": 0.0018,
+      "reward": 2.152442406862974,
+      "reward_std": 1.13260851893574,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.5,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.5069423970999196,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.15625,
+      "rewards/xmlcount_reward_func": 0.8017499931156635,
+      "step": 94
+    },
+    {
+      "completion_length": 263.0625,
+      "epoch": 6.417391304347826,
+      "grad_norm": 6.899081707000732,
+      "kl": 2.110007011331618,
+      "learning_rate": 5.645940686977032e-07,
+      "loss": 0.0021,
+      "reward": 2.49395909672603,
+      "reward_std": 0.8543318175070453,
+      "rewards/concensus_correctness_reward_func": 0.06306250067427754,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.4375,
+      "rewards/question_recreation_reward_func": 0.5518965632654727,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 1.003999998793006,
+      "step": 96
+    },
+    {
+      "completion_length": 256.8125,
+      "epoch": 6.556521739130435,
+      "grad_norm": 18.260860443115234,
+      "kl": 184.23545500356704,
+      "learning_rate": 5.485049264273241e-07,
+      "loss": 0.1842,
+      "reward": 2.550303913652897,
+      "reward_std": 1.0439958362840116,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.5270226672291756,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 0.992031256944756,
+      "step": 98
+    },
+    {
+      "completion_length": 220.03125,
+      "epoch": 6.695652173913043,
+      "grad_norm": 9.59844970703125,
+      "kl": 3.6100155413150787,
+      "learning_rate": 5.323649091872178e-07,
+      "loss": 0.0036,
+      "reward": 2.1622604355216026,
+      "reward_std": 0.7708862484432757,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.5648229457437992,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.109375,
+      "rewards/xmlcount_reward_func": 1.0505624990910292,
+      "step": 100
+    },
+    {
+      "completion_length": 255.9375,
+      "epoch": 6.834782608695652,
+      "grad_norm": 1083.08447265625,
+      "kl": 108.32859648019075,
+      "learning_rate": 5.16190945626678e-07,
+      "loss": 0.1083,
+      "reward": 2.1181236431002617,
+      "reward_std": 1.1967912218533456,
+      "rewards/concensus_correctness_reward_func": 0.023125000298023224,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.5055924141779542,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.15625,
+      "rewards/xmlcount_reward_func": 0.9956562481820583,
+      "step": 102
+    },
+    {
+      "completion_length": 207.9375,
+      "epoch": 6.973913043478261,
+      "grad_norm": 18.834930419921875,
+      "kl": 1.1226458940654993,
+      "learning_rate": 5e-07,
+      "loss": 0.0011,
+      "reward": 2.31644020229578,
+      "reward_std": 0.9884734419174492,
+      "rewards/concensus_correctness_reward_func": 0.0078125,
+      "rewards/consensus_reward_func": 0.1875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.6385339261032641,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.171875,
+      "rewards/xmlcount_reward_func": 0.9982187487185001,
+      "step": 104
+    },
+    {
+      "completion_length": 207.0,
+      "epoch": 7.069565217391304,
+      "grad_norm": 8.474045753479004,
+      "kl": 1.4795172661542892,
+      "learning_rate": 4.838090543733221e-07,
+      "loss": 0.001,
+      "reward": 1.957890353419564,
+      "reward_std": 0.989645315842195,
+      "rewards/concensus_correctness_reward_func": 0.1749090931632302,
+      "rewards/consensus_reward_func": 0.2727272727272727,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.09090909090909091,
+      "rewards/question_recreation_reward_func": 0.29198128560727293,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.1590909090909091,
+      "rewards/xmlcount_reward_func": 0.9682727293534712,
+      "step": 106
+    },
+    {
+      "completion_length": 220.65625,
+      "epoch": 7.208695652173913,
+      "grad_norm": 14.130478858947754,
+      "kl": 2.5718741063028574,
+      "learning_rate": 4.676350908127821e-07,
+      "loss": 0.0026,
+      "reward": 2.538029946386814,
+      "reward_std": 1.10915404278785,
+      "rewards/concensus_correctness_reward_func": 0.07531250268220901,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.4872799450531602,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 1.0223124995827675,
+      "step": 108
+    },
+    {
+      "completion_length": 220.6875,
+      "epoch": 7.3478260869565215,
+      "grad_norm": 73.57449340820312,
+      "kl": 12.105563743039966,
+      "learning_rate": 4.5149507357267597e-07,
+      "loss": 0.0121,
+      "reward": 2.2664614617824554,
+      "reward_std": 0.7128497480880469,
+      "rewards/concensus_correctness_reward_func": 0.06012500077486038,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.6024927124381065,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.078125,
+      "rewards/xmlcount_reward_func": 0.9632187522947788,
+      "step": 110
+    },
+    {
+      "completion_length": 253.90625,
+      "epoch": 7.48695652173913,
+      "grad_norm": 54.51780319213867,
+      "kl": 2.1332995407283306,
+      "learning_rate": 4.354059313022969e-07,
+      "loss": 0.0021,
+      "reward": 2.2279574600979686,
+      "reward_std": 0.8801067136228085,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.6197387175634503,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.078125,
+      "rewards/xmlcount_reward_func": 0.9675937481224537,
+      "step": 112
+    },
+    {
+      "completion_length": 211.25,
+      "epoch": 7.626086956521739,
+      "grad_norm": 159.7990264892578,
+      "kl": 25.76554438099265,
+      "learning_rate": 4.193845392901201e-07,
+      "loss": 0.0258,
+      "reward": 2.2389425486326218,
+      "reward_std": 1.1582523677498102,
+      "rewards/concensus_correctness_reward_func": 0.06200000084936619,
+      "rewards/consensus_reward_func": 0.5625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5203800578601658,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.171875,
+      "rewards/xmlcount_reward_func": 0.8596874997019768,
+      "step": 114
+    },
+    {
+      "completion_length": 210.25,
+      "epoch": 7.765217391304348,
+      "grad_norm": 13.044373512268066,
+      "kl": 2.701926039531827,
+      "learning_rate": 4.0344770176395606e-07,
+      "loss": 0.0027,
+      "reward": 2.635610952973366,
+      "reward_std": 0.8873360179131851,
+      "rewards/concensus_correctness_reward_func": 0.06012500077486038,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.5737046892754734,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.171875,
+      "rewards/xmlcount_reward_func": 1.0799062512814999,
+      "step": 116
+    },
+    {
+      "completion_length": 194.1875,
+      "epoch": 7.904347826086957,
+      "grad_norm": 68.57621765136719,
+      "kl": 11.136024549603462,
+      "learning_rate": 3.8761213426563543e-07,
+      "loss": 0.0111,
+      "reward": 2.058164719492197,
+      "reward_std": 0.8319246040773578,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.5028522202628665,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 0.9771874994039536,
+      "step": 118
+    },
+    {
+      "completion_length": 227.77272727272728,
+      "epoch": 8.0,
+      "grad_norm": 4.235458850860596,
+      "kl": 0.7800953428853642,
+      "learning_rate": 3.718944461187138e-07,
+      "loss": 0.0005,
+      "reward": 2.4683791913769464,
+      "reward_std": 1.1513721336695282,
+      "rewards/concensus_correctness_reward_func": 0.24663636901161887,
+      "rewards/consensus_reward_func": 0.2727272727272727,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.18181818181818182,
+      "rewards/question_recreation_reward_func": 0.5853791890496557,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.18181818181818182,
+      "rewards/xmlcount_reward_func": 1.0,
+      "step": 120
+    },
+    {
+      "completion_length": 206.28125,
+      "epoch": 8.139130434782608,
+      "grad_norm": 7.901040077209473,
+      "kl": 3.598771570250392,
+      "learning_rate": 3.563111230075859e-07,
+      "loss": 0.0036,
+      "reward": 2.611336261034012,
+      "reward_std": 1.0124397999607027,
+      "rewards/concensus_correctness_reward_func": 0.046937499195337296,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.5948674763785675,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.15625,
+      "rewards/xmlcount_reward_func": 1.063281249254942,
+      "step": 122
+    },
+    {
+      "completion_length": 250.65625,
+      "epoch": 8.278260869565218,
+      "grad_norm": 44.01518249511719,
+      "kl": 17.306198243051767,
+      "learning_rate": 3.408785096862782e-07,
+      "loss": 0.0173,
+      "reward": 2.0046014823019505,
+      "reward_std": 0.8732773018418811,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.5254452758817933,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 0.838531251065433,
+      "step": 124
+    },
+    {
+      "completion_length": 225.90625,
+      "epoch": 8.417391304347825,
+      "grad_norm": 9.30581283569336,
+      "kl": 1.4135937942191958,
+      "learning_rate": 3.2561279283505884e-07,
+      "loss": 0.0014,
+      "reward": 2.6753682233393192,
+      "reward_std": 0.8414170006290078,
+      "rewards/concensus_correctness_reward_func": 0.09375,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.5,
+      "rewards/question_recreation_reward_func": 0.5598682607524097,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.171875,
+      "rewards/xmlcount_reward_func": 1.099874995648861,
+      "step": 126
+    },
+    {
+      "completion_length": 218.71875,
+      "epoch": 8.556521739130435,
+      "grad_norm": 20.533634185791016,
+      "kl": 20.78130498714745,
+      "learning_rate": 3.105299840828466e-07,
+      "loss": 0.0208,
+      "reward": 2.386717677116394,
+      "reward_std": 1.0223115284607047,
+      "rewards/concensus_correctness_reward_func": 0.002937499899417162,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.6662176549434662,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.9363125041127205,
+      "step": 128
+    },
+    {
+      "completion_length": 255.84375,
+      "epoch": 8.695652173913043,
+      "grad_norm": 9.14978313446045,
+      "kl": 5.343452694825828,
+      "learning_rate": 2.95645903213222e-07,
+      "loss": 0.0053,
+      "reward": 2.0137369483709335,
+      "reward_std": 0.7903293974231929,
+      "rewards/concensus_correctness_reward_func": 0.06012500077486038,
+      "rewards/consensus_reward_func": 0.5,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.4925806714454666,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.078125,
+      "rewards/xmlcount_reward_func": 0.8829062543809414,
+      "step": 130
+    },
+    {
+      "completion_length": 206.15625,
+      "epoch": 8.834782608695653,
+      "grad_norm": 35.50223922729492,
+      "kl": 4518.369264254812,
+      "learning_rate": 2.8097616157165885e-07,
+      "loss": 4.5184,
+      "reward": 2.349446440115571,
+      "reward_std": 0.8132072143489495,
+      "rewards/concensus_correctness_reward_func": 0.06793750077486038,
+      "rewards/consensus_reward_func": 0.5,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.5124463941901922,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.171875,
+      "rewards/xmlcount_reward_func": 0.9721875041723251,
+      "step": 132
+    },
+    {
+      "completion_length": 283.1875,
+      "epoch": 8.97391304347826,
+      "grad_norm": 11.386509895324707,
+      "kl": 2.30714722070843,
+      "learning_rate": 2.665361456913797e-07,
+      "loss": 0.0023,
+      "reward": 2.2854321654886007,
+      "reward_std": 0.8833654799964279,
+      "rewards/concensus_correctness_reward_func": 0.06012500077486038,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.5298071521683596,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.1875,
+      "rewards/xmlcount_reward_func": 1.0079999975860119,
+      "step": 134
+    },
+    {
+      "completion_length": 276.54545454545456,
+      "epoch": 9.069565217391304,
+      "grad_norm": 36.2149772644043,
+      "kl": 8.192625154148448,
+      "learning_rate": 2.523410011550064e-07,
+      "loss": 0.0056,
+      "reward": 1.7320992946624756,
+      "reward_std": 0.8099311292171478,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.2727272727272727,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.4210083870725198,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.11363636363636363,
+      "rewards/xmlcount_reward_func": 0.9247272664850409,
+      "step": 136
+    },
+    {
+      "completion_length": 291.78125,
+      "epoch": 9.208695652173914,
+      "grad_norm": 7.384568214416504,
+      "kl": 79.57251481059939,
+      "learning_rate": 2.3840561670893495e-07,
+      "loss": 0.0796,
+      "reward": 2.1450905594974756,
+      "reward_std": 0.9664679318666458,
+      "rewards/concensus_correctness_reward_func": 0.06012500077486038,
+      "rewards/consensus_reward_func": 0.5,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5948093222687021,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 0.833906251937151,
+      "step": 138
+    },
+    {
+      "completion_length": 287.0,
+      "epoch": 9.347826086956522,
+      "grad_norm": 7.677389621734619,
+      "kl": 1.0843814071267843,
+      "learning_rate": 2.247446086470982e-07,
+      "loss": 0.0011,
+      "reward": 2.9537404403090477,
+      "reward_std": 1.031734160031192,
+      "rewards/concensus_correctness_reward_func": 0.12812500167638063,
+      "rewards/consensus_reward_func": 0.6875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.6615529721602798,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.15625,
+      "rewards/xmlcount_reward_func": 1.0703125,
+      "step": 140
+    },
+    {
+      "completion_length": 195.75,
+      "epoch": 9.486956521739131,
+      "grad_norm": 14.392241477966309,
+      "kl": 1.338534751906991,
+      "learning_rate": 2.113723054804904e-07,
+      "loss": 0.0013,
+      "reward": 2.5440589003264904,
+      "reward_std": 0.5958110167412087,
+      "rewards/concensus_correctness_reward_func": 0.002937499899417162,
+      "rewards/consensus_reward_func": 0.5,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.48871511360630393,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.1875,
+      "rewards/xmlcount_reward_func": 1.1149062551558018,
+      "step": 142
+    },
+    {
+      "completion_length": 280.0625,
+      "epoch": 9.626086956521739,
+      "grad_norm": 9.929582595825195,
+      "kl": 6.2964247446507215,
+      "learning_rate": 1.9830273290853766e-07,
+      "loss": 0.0063,
+      "reward": 2.0625064894557,
+      "reward_std": 0.8793220520019531,
+      "rewards/concensus_correctness_reward_func": 0.1172500029206276,
+      "rewards/consensus_reward_func": 0.0625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.6093190036481246,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 1.0078125,
+      "step": 144
+    },
+    {
+      "completion_length": 230.53125,
+      "epoch": 9.765217391304347,
+      "grad_norm": 93.21333312988281,
+      "kl": 10.086717769503593,
+      "learning_rate": 1.8554959910807772e-07,
+      "loss": 0.0101,
+      "reward": 2.2185809947550297,
+      "reward_std": 1.1026300959056243,
+      "rewards/concensus_correctness_reward_func": 0.145687498152256,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.5238622860051692,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 0.9709062464535236,
+      "step": 146
+    },
+    {
+      "completion_length": 245.90625,
+      "epoch": 9.904347826086957,
+      "grad_norm": 7.6316728591918945,
+      "kl": 2.9326881784945726,
+      "learning_rate": 1.7312628035537386e-07,
+      "loss": 0.0029,
+      "reward": 2.5975780775770545,
+      "reward_std": 1.2725993720814586,
+      "rewards/concensus_correctness_reward_func": 0.0832500010728836,
+      "rewards/consensus_reward_func": 0.5,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.4375,
+      "rewards/question_recreation_reward_func": 0.6529530733823776,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 0.8301250021904707,
+      "step": 148
+    },
+    {
+      "completion_length": 238.9090909090909,
+      "epoch": 10.0,
+      "grad_norm": 3.3495516777038574,
+      "kl": 1.7505317547104575,
+      "learning_rate": 1.6104580699624837e-07,
+      "loss": 0.0012,
+      "reward": 2.4565178697759453,
+      "reward_std": 0.8326517287218436,
+      "rewards/concensus_correctness_reward_func": 0.06836363402279941,
+      "rewards/consensus_reward_func": 0.6363636363636364,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.604063347320665,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.13636363636363635,
+      "rewards/xmlcount_reward_func": 1.0113636363636365,
+      "step": 150
+    },
+    {
+      "completion_length": 223.6875,
+      "epoch": 10.139130434782608,
+      "grad_norm": 8.917014122009277,
+      "kl": 1036.1936433017254,
+      "learning_rate": 1.493208497790504e-07,
+      "loss": 1.0362,
+      "reward": 1.8704981822520494,
+      "reward_std": 0.8263477487489581,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.1875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.4501856912393123,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 0.8265624977648258,
+      "step": 152
+    },
+    {
+      "completion_length": 304.3125,
+      "epoch": 10.278260869565218,
+      "grad_norm": 109.25184631347656,
+      "kl": 6.692760448902845,
+      "learning_rate": 1.3796370656478934e-07,
+      "loss": 0.0067,
+      "reward": 1.8072622884064913,
+      "reward_std": 1.073743922635913,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.40573102713096887,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.109375,
+      "rewards/xmlcount_reward_func": 0.7921562530100346,
+      "step": 154
+    },
+    {
+      "completion_length": 196.1875,
+      "epoch": 10.417391304347825,
+      "grad_norm": 8.076583862304688,
+      "kl": 21.29381263256073,
+      "learning_rate": 1.2698628942837697e-07,
+      "loss": 0.0213,
+      "reward": 2.1946525387465954,
+      "reward_std": 0.7001782967709005,
+      "rewards/concensus_correctness_reward_func": 0.0078125,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.5274649991188198,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 1.003125000745058,
+      "step": 156
+    },
+    {
+      "completion_length": 294.90625,
+      "epoch": 10.556521739130435,
+      "grad_norm": 10.061820983886719,
+      "kl": 2.001404521986842,
+      "learning_rate": 1.1640011216450691e-07,
+      "loss": 0.002,
+      "reward": 2.073508506640792,
+      "reward_std": 1.356637116521597,
+      "rewards/concensus_correctness_reward_func": 0.09112500119954348,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.363852248119656,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 0.8685312494635582,
+      "step": 158
+    },
+    {
+      "completion_length": 238.5625,
+      "epoch": 10.695652173913043,
+      "grad_norm": 7.213126182556152,
+      "kl": 1.908262014389038,
+      "learning_rate": 1.0621627821127288e-07,
+      "loss": 0.0019,
+      "reward": 2.791898600757122,
+      "reward_std": 1.6709155699936673,
+      "rewards/concensus_correctness_reward_func": 0.1901874989271164,
+      "rewards/consensus_reward_func": 0.625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.48136736266314983,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 1.0578437522053719,
+      "step": 160
+    },
+    {
+      "completion_length": 231.71875,
+      "epoch": 10.834782608695653,
+      "grad_norm": 34.76180648803711,
+      "kl": 5.669580914080143,
+      "learning_rate": 9.644546900419531e-08,
+      "loss": 0.0057,
+      "reward": 1.957098614424467,
+      "reward_std": 0.7321692104451358,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.4589111144305207,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.078125,
+      "rewards/xmlcount_reward_func": 0.9200624991208315,
+      "step": 162
+    },
+    {
+      "completion_length": 274.625,
+      "epoch": 10.97391304347826,
+      "grad_norm": 8.54128360748291,
+      "kl": 3.205599080771208,
+      "learning_rate": 8.70979327728718e-08,
+      "loss": 0.0032,
+      "reward": 2.795065075159073,
+      "reward_std": 1.1369968629442155,
+      "rewards/concensus_correctness_reward_func": 0.12025000154972076,
+      "rewards/consensus_reward_func": 0.5625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.6334713499527425,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 1.0100937485694885,
+      "step": 164
+    },
+    {
+      "completion_length": 168.27272727272728,
+      "epoch": 11.069565217391304,
+      "grad_norm": 19.19721794128418,
+      "kl": 3.619182462042028,
+      "learning_rate": 7.81834737919978e-08,
+      "loss": 0.0025,
+      "reward": 3.018909389322454,
+      "reward_std": 0.7642647468230941,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.8181818181818182,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.18181818181818182,
+      "rewards/question_recreation_reward_func": 0.6259548024020412,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.2727272727272727,
+      "rewards/xmlcount_reward_func": 1.1202272718602961,
+      "step": 166
+    },
+    {
+      "completion_length": 214.65625,
+      "epoch": 11.208695652173914,
+      "grad_norm": 22.970157623291016,
+      "kl": 2.239942977204919,
+      "learning_rate": 6.971144209803736e-08,
+      "loss": 0.0022,
+      "reward": 1.9759991057217121,
+      "reward_std": 0.8322795821586624,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5072491308674216,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 1.015625,
+      "step": 168
+    },
+    {
+      "completion_length": 268.09375,
+      "epoch": 11.347826086956522,
+      "grad_norm": 8.616852760314941,
+      "kl": 1.2630443252855912,
+      "learning_rate": 6.16907236823262e-08,
+      "loss": 0.0013,
+      "reward": 1.666411878541112,
+      "reward_std": 0.861880192087483,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.25,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.40978686296148226,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 0.6941249966621399,
+      "step": 170
+    },
+    {
+      "completion_length": 218.5,
+      "epoch": 11.486956521739131,
+      "grad_norm": 11.935127258300781,
+      "kl": 3.4988537207245827,
+      "learning_rate": 5.412973117089287e-08,
+      "loss": 0.0035,
+      "reward": 2.7520089522004128,
+      "reward_std": 1.139378215186298,
+      "rewards/concensus_correctness_reward_func": 0.05412500072270632,
+      "rewards/consensus_reward_func": 0.625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.7122902190312743,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 0.9230937529355288,
+      "step": 172
+    },
+    {
+      "completion_length": 176.59375,
+      "epoch": 11.626086956521739,
+      "grad_norm": 15.085416793823242,
+      "kl": 2.039413772523403,
+      "learning_rate": 4.703639500077655e-08,
+      "loss": 0.002,
+      "reward": 2.7363155260682106,
+      "reward_std": 0.9747829738153087,
+      "rewards/concensus_correctness_reward_func": 0.22962500154972076,
+      "rewards/consensus_reward_func": 0.8125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5008780010975897,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 1.005812507122755,
+      "step": 174
+    },
+    {
+      "completion_length": 245.84375,
+      "epoch": 11.765217391304347,
+      "grad_norm": 8.771071434020996,
+      "kl": 0.9153302293270826,
+      "learning_rate": 4.041815510209395e-08,
+      "loss": 0.0009,
+      "reward": 2.9970211051404476,
+      "reward_std": 0.8632267946668435,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.875,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.5793023159494624,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.25,
+      "rewards/xmlcount_reward_func": 1.0427187532186508,
+      "step": 176
+    },
+    {
+      "completion_length": 181.96875,
+      "epoch": 11.904347826086957,
+      "grad_norm": 6.449145793914795,
+      "kl": 1.2241215463727713,
+      "learning_rate": 3.4281953094578875e-08,
+      "loss": 0.0012,
+      "reward": 2.4653413705527782,
+      "reward_std": 0.5194937253836542,
+      "rewards/concensus_correctness_reward_func": 0.009437499567866325,
+      "rewards/consensus_reward_func": 0.5625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.493560123257339,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 1.0248437486588955,
+      "step": 178
+    },
+    {
+      "completion_length": 248.0,
+      "epoch": 12.0,
+      "grad_norm": 4.065160274505615,
+      "kl": 0.9604931798848239,
+      "learning_rate": 2.8634225006782864e-08,
+      "loss": 0.0007,
+      "reward": 2.7070359858599575,
+      "reward_std": 1.2332754209637642,
+      "rewards/concensus_correctness_reward_func": 0.18627272952686658,
+      "rewards/consensus_reward_func": 0.6363636363636364,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.09090909090909091,
+      "rewards/question_recreation_reward_func": 0.5386723849753087,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.11363636363636363,
+      "rewards/xmlcount_reward_func": 1.1411818157542835,
+      "step": 180
+    },
+    {
+      "completion_length": 284.875,
+      "epoch": 12.139130434782608,
+      "grad_norm": 6.747330188751221,
+      "kl": 1.845324163325131,
+      "learning_rate": 2.348089452556956e-08,
+      "loss": 0.0018,
+      "reward": 2.8983127176761627,
+      "reward_std": 1.034126427024603,
+      "rewards/concensus_correctness_reward_func": 0.09425000194460154,
+      "rewards/consensus_reward_func": 0.8125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.6407502153888345,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.125,
+      "rewards/xmlcount_reward_func": 1.0383125022053719,
+      "step": 182
+    },
+    {
+      "completion_length": 195.59375,
+      "epoch": 12.278260869565218,
+      "grad_norm": 8.408949851989746,
+      "kl": 1.878015061840415,
+      "learning_rate": 1.882736678298491e-08,
+      "loss": 0.0019,
+      "reward": 1.9930787160992622,
+      "reward_std": 0.7291824659332633,
+      "rewards/concensus_correctness_reward_func": 0.07581249997019768,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.48314121598377824,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.078125,
+      "rewards/xmlcount_reward_func": 0.9810000024735928,
+      "step": 184
+    },
+    {
+      "completion_length": 263.0,
+      "epoch": 12.417391304347825,
+      "grad_norm": 14483.93359375,
+      "kl": 475.0545411463827,
+      "learning_rate": 1.4678522687020412e-08,
+      "loss": 0.4751,
+      "reward": 2.7335060872137547,
+      "reward_std": 0.894551997538656,
+      "rewards/concensus_correctness_reward_func": 0.12025000154972076,
+      "rewards/consensus_reward_func": 0.75,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.46903733629733324,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 0.9879687428474426,
+      "step": 186
+    },
+    {
+      "completion_length": 244.96875,
+      "epoch": 12.556521739130435,
+      "grad_norm": 6.763554573059082,
+      "kl": 1.7115775179117918,
+      "learning_rate": 1.1038713802214717e-08,
+      "loss": 0.0017,
+      "reward": 2.217247996479273,
+      "reward_std": 1.002159099560231,
+      "rewards/concensus_correctness_reward_func": 0.06775000039488077,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.6073104892857373,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.109375,
+      "rewards/xmlcount_reward_func": 1.057812500745058,
+      "step": 188
+    },
+    {
+      "completion_length": 307.34375,
+      "epoch": 12.695652173913043,
+      "grad_norm": 22.309476852416992,
+      "kl": 2.2677318193018436,
+      "learning_rate": 7.91175778546288e-09,
+      "loss": 0.0023,
+      "reward": 2.050308357924223,
+      "reward_std": 1.076811763457954,
+      "rewards/concensus_correctness_reward_func": 0.023499999195337296,
+      "rewards/consensus_reward_func": 0.375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.3125,
+      "rewards/question_recreation_reward_func": 0.5594645974924788,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 0.6860937541350722,
+      "step": 190
+    },
+    {
+      "completion_length": 263.90625,
+      "epoch": 12.834782608695653,
+      "grad_norm": 8.240316390991211,
+      "kl": 14.615491598844528,
+      "learning_rate": 5.3009343818219975e-09,
+      "loss": 0.0146,
+      "reward": 2.2848374396562576,
+      "reward_std": 1.0299444798147306,
+      "rewards/concensus_correctness_reward_func": 0.023499999195337296,
+      "rewards/consensus_reward_func": 0.3125,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.5453686523251235,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.140625,
+      "rewards/xmlcount_reward_func": 0.8878437541425228,
+      "step": 192
+    },
+    {
+      "completion_length": 181.125,
+      "epoch": 12.97391304347826,
+      "grad_norm": 10.09430980682373,
+      "kl": 2.5310205705463886,
+      "learning_rate": 3.2089819845111944e-09,
+      "loss": 0.0025,
+      "reward": 2.4083995819091797,
+      "reward_std": 0.6754440142103704,
+      "rewards/concensus_correctness_reward_func": 0.030937500298023224,
+      "rewards/consensus_reward_func": 0.5625,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.5698058666894212,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.109375,
+      "rewards/xmlcount_reward_func": 1.0732812508940697,
+      "step": 194
+    },
+    {
+      "completion_length": 252.77272727272728,
+      "epoch": 13.069565217391304,
+      "grad_norm": 34.62232971191406,
+      "kl": 4.217680139975115,
+      "learning_rate": 1.638094762715314e-09,
+      "loss": 0.0029,
+      "reward": 1.9441369067538867,
+      "reward_std": 0.6365875859152187,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.18181818181818182,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.36363636363636365,
+      "rewards/question_recreation_reward_func": 0.5173641754822298,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.022727272727272728,
+      "rewards/xmlcount_reward_func": 0.8585909144444899,
+      "step": 196
+    },
+    {
+      "completion_length": 207.46875,
+      "epoch": 13.208695652173914,
+      "grad_norm": 9.698592185974121,
+      "kl": 5.665437141433358,
+      "learning_rate": 5.899203602046654e-10,
+      "loss": 0.0057,
+      "reward": 2.258667390793562,
+      "reward_std": 0.8544780421070755,
+      "rewards/concensus_correctness_reward_func": 0.002937499899417162,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.555886154237669,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.046875,
+      "rewards/xmlcount_reward_func": 0.9654687494039536,
+      "step": 198
+    },
+    {
+      "completion_length": 223.0625,
+      "epoch": 13.347826086956522,
+      "grad_norm": 9.346601486206055,
+      "kl": 1.620649179443717,
+      "learning_rate": 6.555816718389895e-11,
+      "loss": 0.0016,
+      "reward": 2.1773801445961,
+      "reward_std": 1.0183756935875863,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.4375,
+      "rewards/cumulative_reward_2": 0.0,
+      "rewards/final_correctness_reward_func": 0.375,
+      "rewards/question_recreation_reward_func": 0.4039426180534065,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.09375,
+      "rewards/xmlcount_reward_func": 0.8671875,
+      "step": 200
+    },
+    {
+      "epoch": 13.347826086956522,
+      "step": 200,
+      "total_flos": 0.0,
+      "train_loss": 0.06702521831626655,
+      "train_runtime": 3029.6015,
+      "train_samples_per_second": 1.056,
+      "train_steps_per_second": 0.066
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 200,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 15,
+  "save_steps": 25,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "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:d6829c076bda83803fc3ec9735d3b6fcf0ba3ba60bef27da15a48404b8ffbb18
+size 5944