Add files using upload-large-folder tool

.gitattributes

@@ -59,3 +59,10 @@ Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1200/tokenizer.json filter=lfs d
 Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1700/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-900/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+llama3b-GRPO-Code-vllm/checkpoint-200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-700/tokenizer.json filter=lfs diff=lfs merge=lfs -text

Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-700/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
+}

Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-700/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-0.5B-Instruct",
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-700/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": "4.48.2"
+}

Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-700/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ca74a99676c40cc90e22f342ab38a573fba6b6e0c9a26aba378c13a08ec992f
+size 14448

Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-700/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f68a37892a1b445d21bb35cc10bf7a058a6f9ec8c363f5ed156ff4f49d90fb6
+size 14512

Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-700/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:893d4b8b1302e161f9c549f4f9074ed54107bc52757ec993a56b789ee94ddea5
+size 1064

Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-700/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "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": "<|im_end|>"
+}

Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-700/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63a2951d5edfa5cc0a2346ef872f8c77a2920274cfc3b503b04e3799104dee80
+size 11422060

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/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
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "./outputs/Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-800",
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.49.0.dev0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/generation_config.json

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

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4a658b8ca13664e325ac79509473bc459e83fb410b27e6f0fa00f21418667a4
+size 14448

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71d2161c9233860a0c55de4fa68dfcf16cdf4af9234c3d44fc611fcd896d1189
+size 1064

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "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": "<|im_end|>"
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63a2951d5edfa5cc0a2346ef872f8c77a2920274cfc3b503b04e3799104dee80
+size 11422060

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/tokenizer_config.json

@@ -0,0 +1,212 @@
+{
+  "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": {},
+  "max_length": null,
+  "model_max_length": 131072,
+  "pad_to_multiple_of": null,
+  "pad_token": "<|im_end|>",
+  "pad_token_type_id": 0,
+  "padding_side": "left",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/trainer_state.json

@@ -0,0 +1,1633 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.02675943270002676,
+  "eval_steps": 500,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 240.25,
+      "epoch": 0.0002675943270002676,
+      "grad_norm": 6.21875,
+      "kl": 0.0,
+      "learning_rate": 5.3475935828877005e-08,
+      "loss": -0.0,
+      "reward": 1.683333396911621,
+      "reward_std": 0.4003768861293793,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.2916666865348816,
+      "rewards/code_quality_reward_func": 0.3166666626930237,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.2666666805744171,
+      "step": 1
+    },
+    {
+      "completion_length": 332.3333435058594,
+      "epoch": 0.0005351886540005352,
+      "grad_norm": 4.40625,
+      "kl": 0.0,
+      "learning_rate": 1.0695187165775401e-07,
+      "loss": 0.0,
+      "reward": 1.2833563089370728,
+      "reward_std": 0.31695565581321716,
+      "rewards/basic_format_reward_func": 0.1666666716337204,
+      "rewards/code_execution_reward": 0.266689658164978,
+      "rewards/code_quality_reward_func": 0.27500003576278687,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.18333333730697632,
+      "step": 2
+    },
+    {
+      "completion_length": 270.25,
+      "epoch": 0.0008027829810008028,
+      "grad_norm": 5.125,
+      "kl": 0.0007383245974779129,
+      "learning_rate": 1.6042780748663104e-07,
+      "loss": 0.0,
+      "reward": 1.6968612670898438,
+      "reward_std": 0.3620304465293884,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.32186126708984375,
+      "rewards/code_quality_reward_func": 0.34166666865348816,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 3
+    },
+    {
+      "completion_length": 324.0,
+      "epoch": 0.0010703773080010704,
+      "grad_norm": 4.3125,
+      "kl": 0.0006790084880776703,
+      "learning_rate": 2.1390374331550802e-07,
+      "loss": 0.0,
+      "reward": 1.723926305770874,
+      "reward_std": 0.5276123881340027,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.42392635345458984,
+      "rewards/code_quality_reward_func": 0.28333336114883423,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 4
+    },
+    {
+      "completion_length": 200.0,
+      "epoch": 0.001337971635001338,
+      "grad_norm": 5.5,
+      "kl": 0.0009003459708765149,
+      "learning_rate": 2.6737967914438503e-07,
+      "loss": 0.0,
+      "reward": 1.6696914434432983,
+      "reward_std": 0.7859481573104858,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.4113580584526062,
+      "rewards/code_quality_reward_func": 0.28333336114883423,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.24166667461395264,
+      "step": 5
+    },
+    {
+      "completion_length": 284.5833435058594,
+      "epoch": 0.0016055659620016055,
+      "grad_norm": 4.96875,
+      "kl": 0.0007309447391889989,
+      "learning_rate": 3.208556149732621e-07,
+      "loss": 0.0,
+      "reward": 1.6401863098144531,
+      "reward_std": 0.38958868384361267,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.27351975440979004,
+      "rewards/code_quality_reward_func": 0.3333333432674408,
+      "rewards/reasoning_quality_reward_func": 0.34166666865348816,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 6
+    },
+    {
+      "completion_length": 292.16668701171875,
+      "epoch": 0.0018731602890018732,
+      "grad_norm": 6.875,
+      "kl": 0.0008700569160282612,
+      "learning_rate": 3.7433155080213904e-07,
+      "loss": 0.0,
+      "reward": 1.718916893005371,
+      "reward_std": 0.8036686778068542,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.4355836808681488,
+      "rewards/code_quality_reward_func": 0.2916666865348816,
+      "rewards/reasoning_quality_reward_func": 0.3583333492279053,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 7
+    },
+    {
+      "completion_length": 216.1666717529297,
+      "epoch": 0.0021407546160021407,
+      "grad_norm": 4.65625,
+      "kl": 0.0005313638830557466,
+      "learning_rate": 4.2780748663101604e-07,
+      "loss": 0.0,
+      "reward": 2.3377904891967773,
+      "reward_std": 1.373246192932129,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 1.046123743057251,
+      "rewards/code_quality_reward_func": 0.28333336114883423,
+      "rewards/reasoning_quality_reward_func": 0.36666667461395264,
+      "rewards/xml_structure_reward_func": 0.2666667103767395,
+      "step": 8
+    },
+    {
+      "completion_length": 291.25,
+      "epoch": 0.002408348943002408,
+      "grad_norm": 5.34375,
+      "kl": 0.000748022343032062,
+      "learning_rate": 4.812834224598931e-07,
+      "loss": 0.0,
+      "reward": 1.5542452335357666,
+      "reward_std": 0.6039680242538452,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.3625785708427429,
+      "rewards/code_quality_reward_func": 0.23333333432674408,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.23333334922790527,
+      "step": 9
+    },
+    {
+      "completion_length": 241.08334350585938,
+      "epoch": 0.002675943270002676,
+      "grad_norm": 5.46875,
+      "kl": 0.0007344354526139796,
+      "learning_rate": 5.347593582887701e-07,
+      "loss": 0.0,
+      "reward": 1.7666667699813843,
+      "reward_std": 0.24568384885787964,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.32500001788139343,
+      "rewards/code_quality_reward_func": 0.3083333373069763,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.2916666865348816,
+      "step": 10
+    },
+    {
+      "completion_length": 243.08334350585938,
+      "epoch": 0.0029435375970029436,
+      "grad_norm": 5.59375,
+      "kl": 0.0006488829385489225,
+      "learning_rate": 5.882352941176471e-07,
+      "loss": 0.0,
+      "reward": 1.9167588949203491,
+      "reward_std": 0.7602280378341675,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.6334255337715149,
+      "rewards/code_quality_reward_func": 0.25833332538604736,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 11
+    },
+    {
+      "completion_length": 217.75,
+      "epoch": 0.003211131924003211,
+      "grad_norm": 6.59375,
+      "kl": 0.0010273342486470938,
+      "learning_rate": 6.417112299465242e-07,
+      "loss": 0.0,
+      "reward": 1.6952461004257202,
+      "reward_std": 0.43511244654655457,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.3369128108024597,
+      "rewards/code_quality_reward_func": 0.28333336114883423,
+      "rewards/reasoning_quality_reward_func": 0.3916666507720947,
+      "rewards/xml_structure_reward_func": 0.2666667103767395,
+      "step": 12
+    },
+    {
+      "completion_length": 241.33334350585938,
+      "epoch": 0.0034787262510034785,
+      "grad_norm": 4.5625,
+      "kl": 0.0006450031069107354,
+      "learning_rate": 6.951871657754011e-07,
+      "loss": 0.0,
+      "reward": 2.3087432384490967,
+      "reward_std": 1.2053987979888916,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.8837432265281677,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.3583333492279053,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 13
+    },
+    {
+      "completion_length": 224.25,
+      "epoch": 0.0037463205780037465,
+      "grad_norm": 5.25,
+      "kl": 0.0006655638571828604,
+      "learning_rate": 7.486631016042781e-07,
+      "loss": 0.0,
+      "reward": 2.016702890396118,
+      "reward_std": 0.8577584624290466,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.7167028188705444,
+      "rewards/code_quality_reward_func": 0.30000001192092896,
+      "rewards/reasoning_quality_reward_func": 0.36666667461395264,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 14
+    },
+    {
+      "completion_length": 278.91668701171875,
+      "epoch": 0.004013914905004014,
+      "grad_norm": 6.1875,
+      "kl": 0.0008740181219764054,
+      "learning_rate": 8.021390374331551e-07,
+      "loss": 0.0,
+      "reward": 1.3916771411895752,
+      "reward_std": 0.5112754106521606,
+      "rewards/basic_format_reward_func": 0.2916666865348816,
+      "rewards/code_execution_reward": 0.24167709052562714,
+      "rewards/code_quality_reward_func": 0.23333334922790527,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.23333333432674408,
+      "step": 15
+    },
+    {
+      "completion_length": 223.5,
+      "epoch": 0.004281509232004281,
+      "grad_norm": 5.5625,
+      "kl": 0.0007978550856932998,
+      "learning_rate": 8.556149732620321e-07,
+      "loss": 0.0,
+      "reward": 1.9002037048339844,
+      "reward_std": 0.6648212671279907,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.5418704152107239,
+      "rewards/code_quality_reward_func": 0.28333336114883423,
+      "rewards/reasoning_quality_reward_func": 0.3916666507720947,
+      "rewards/xml_structure_reward_func": 0.2666667103767395,
+      "step": 16
+    },
+    {
+      "completion_length": 326.41668701171875,
+      "epoch": 0.004549103559004549,
+      "grad_norm": 4.375,
+      "kl": 0.0007469221018254757,
+      "learning_rate": 9.090909090909091e-07,
+      "loss": 0.0,
+      "reward": 1.2916667461395264,
+      "reward_std": 0.49194058775901794,
+      "rewards/basic_format_reward_func": 0.2083333432674408,
+      "rewards/code_execution_reward": 0.25833332538604736,
+      "rewards/code_quality_reward_func": 0.22500000894069672,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.20000001788139343,
+      "step": 17
+    },
+    {
+      "completion_length": 246.5,
+      "epoch": 0.004816697886004816,
+      "grad_norm": 6.46875,
+      "kl": 0.0007637516828253865,
+      "learning_rate": 9.625668449197862e-07,
+      "loss": 0.0,
+      "reward": 1.6097962856292725,
+      "reward_std": 0.5806087255477905,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.3681294620037079,
+      "rewards/code_quality_reward_func": 0.2666666805744171,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.25,
+      "step": 18
+    },
+    {
+      "completion_length": 225.6666717529297,
+      "epoch": 0.005084292213005084,
+      "grad_norm": 6.15625,
+      "kl": 0.0007218674290925264,
+      "learning_rate": 1.0160427807486633e-06,
+      "loss": 0.0,
+      "reward": 1.367079734802246,
+      "reward_std": 0.3885188698768616,
+      "rewards/basic_format_reward_func": 0.2083333432674408,
+      "rewards/code_execution_reward": 0.24207979440689087,
+      "rewards/code_quality_reward_func": 0.32499998807907104,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.20000001788139343,
+      "step": 19
+    },
+    {
+      "completion_length": 283.66668701171875,
+      "epoch": 0.005351886540005352,
+      "grad_norm": 4.90625,
+      "kl": 0.0008003411930985749,
+      "learning_rate": 1.0695187165775401e-06,
+      "loss": 0.0,
+      "reward": 1.5653541088104248,
+      "reward_std": 0.4483262300491333,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.323687344789505,
+      "rewards/code_quality_reward_func": 0.28333336114883423,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.24166667461395264,
+      "step": 20
+    },
+    {
+      "completion_length": 268.41668701171875,
+      "epoch": 0.00561948086700562,
+      "grad_norm": 4.5625,
+      "kl": 0.0007850835099816322,
+      "learning_rate": 1.1229946524064172e-06,
+      "loss": 0.0,
+      "reward": 1.8270270824432373,
+      "reward_std": 0.7158061861991882,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.5186935663223267,
+      "rewards/code_quality_reward_func": 0.2916666865348816,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 21
+    },
+    {
+      "completion_length": 247.58334350585938,
+      "epoch": 0.005887075194005887,
+      "grad_norm": 7.3125,
+      "kl": 0.0008080464322119951,
+      "learning_rate": 1.1764705882352942e-06,
+      "loss": 0.0,
+      "reward": 1.5754141807556152,
+      "reward_std": 0.32768064737319946,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.17541414499282837,
+      "rewards/code_quality_reward_func": 0.3499999940395355,
+      "rewards/reasoning_quality_reward_func": 0.3499999940395355,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 22
+    },
+    {
+      "completion_length": 252.08334350585938,
+      "epoch": 0.006154669521006155,
+      "grad_norm": 6.53125,
+      "kl": 0.0009499495499767363,
+      "learning_rate": 1.2299465240641713e-06,
+      "loss": 0.0,
+      "reward": 1.2697193622589111,
+      "reward_std": 0.40648603439331055,
+      "rewards/basic_format_reward_func": 0.25,
+      "rewards/code_execution_reward": 0.1780526041984558,
+      "rewards/code_quality_reward_func": 0.27500003576278687,
+      "rewards/reasoning_quality_reward_func": 0.3583333492279053,
+      "rewards/xml_structure_reward_func": 0.2083333432674408,
+      "step": 23
+    },
+    {
+      "completion_length": 280.5,
+      "epoch": 0.006422263848006422,
+      "grad_norm": 4.8125,
+      "kl": 0.00072054925840348,
+      "learning_rate": 1.2834224598930483e-06,
+      "loss": 0.0,
+      "reward": 1.7976078987121582,
+      "reward_std": 0.40162503719329834,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.3309413194656372,
+      "rewards/code_quality_reward_func": 0.3333333432674408,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 24
+    },
+    {
+      "completion_length": 187.4166717529297,
+      "epoch": 0.00668985817500669,
+      "grad_norm": 6.71875,
+      "kl": 0.001055043889209628,
+      "learning_rate": 1.3368983957219254e-06,
+      "loss": 0.0,
+      "reward": 1.9250602722167969,
+      "reward_std": 0.7650420069694519,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.6500604748725891,
+      "rewards/code_quality_reward_func": 0.34166666865348816,
+      "rewards/reasoning_quality_reward_func": 0.3499999940395355,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 25
+    },
+    {
+      "completion_length": 223.1666717529297,
+      "epoch": 0.006957452502006957,
+      "grad_norm": 5.8125,
+      "kl": 0.0008547938195988536,
+      "learning_rate": 1.3903743315508022e-06,
+      "loss": 0.0,
+      "reward": 1.5859073400497437,
+      "reward_std": 0.8232795596122742,
+      "rewards/basic_format_reward_func": 0.2916666865348816,
+      "rewards/code_execution_reward": 0.38590726256370544,
+      "rewards/code_quality_reward_func": 0.2916666865348816,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.23333333432674408,
+      "step": 26
+    },
+    {
+      "completion_length": 251.58334350585938,
+      "epoch": 0.0072250468290072254,
+      "grad_norm": 5.03125,
+      "kl": 0.0007728833588771522,
+      "learning_rate": 1.4438502673796793e-06,
+      "loss": 0.0,
+      "reward": 1.491668701171875,
+      "reward_std": 0.44286617636680603,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.19166871905326843,
+      "rewards/code_quality_reward_func": 0.27500003576278687,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 27
+    },
+    {
+      "completion_length": 372.0,
+      "epoch": 0.007492641156007493,
+      "grad_norm": 4.6875,
+      "kl": 0.001101998845115304,
+      "learning_rate": 1.4973262032085562e-06,
+      "loss": 0.0,
+      "reward": 1.5583410263061523,
+      "reward_std": 0.4496191442012787,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.23334108293056488,
+      "rewards/code_quality_reward_func": 0.3083333373069763,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 28
+    },
+    {
+      "completion_length": 297.75,
+      "epoch": 0.00776023548300776,
+      "grad_norm": 4.46875,
+      "kl": 0.0007874669972807169,
+      "learning_rate": 1.5508021390374334e-06,
+      "loss": 0.0,
+      "reward": 1.6197443008422852,
+      "reward_std": 0.2738063931465149,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.1864110678434372,
+      "rewards/code_quality_reward_func": 0.3583333492279053,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 29
+    },
+    {
+      "completion_length": 321.75,
+      "epoch": 0.008027829810008028,
+      "grad_norm": 4.1875,
+      "kl": 0.0007515697507187724,
+      "learning_rate": 1.6042780748663103e-06,
+      "loss": 0.0,
+      "reward": 1.5833333730697632,
+      "reward_std": 0.37617573142051697,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.22499999403953552,
+      "rewards/code_quality_reward_func": 0.2666666805744171,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 30
+    },
+    {
+      "completion_length": 251.58334350585938,
+      "epoch": 0.008295424137008296,
+      "grad_norm": 6.5,
+      "kl": 0.0012028946075588465,
+      "learning_rate": 1.6577540106951873e-06,
+      "loss": 0.0,
+      "reward": 1.6526631116867065,
+      "reward_std": 0.5531408786773682,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.3693297803401947,
+      "rewards/code_quality_reward_func": 0.34166669845581055,
+      "rewards/reasoning_quality_reward_func": 0.3500000238418579,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 31
+    },
+    {
+      "completion_length": 303.91668701171875,
+      "epoch": 0.008563018464008563,
+      "grad_norm": 4.46875,
+      "kl": 0.0007428948301821947,
+      "learning_rate": 1.7112299465240642e-06,
+      "loss": 0.0,
+      "reward": 1.7810885906219482,
+      "reward_std": 0.43417853116989136,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.3227551579475403,
+      "rewards/code_quality_reward_func": 0.32499998807907104,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 32
+    },
+    {
+      "completion_length": 294.16668701171875,
+      "epoch": 0.008830612791008831,
+      "grad_norm": 4.5625,
+      "kl": 0.0007438965840265155,
+      "learning_rate": 1.7647058823529414e-06,
+      "loss": 0.0,
+      "reward": 1.663476824760437,
+      "reward_std": 0.34124574065208435,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.21347686648368835,
+      "rewards/code_quality_reward_func": 0.3083333373069763,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 33
+    },
+    {
+      "completion_length": 246.33334350585938,
+      "epoch": 0.009098207118009098,
+      "grad_norm": 5.3125,
+      "kl": 0.0010945609537884593,
+      "learning_rate": 1.8181818181818183e-06,
+      "loss": 0.0,
+      "reward": 1.6503428220748901,
+      "reward_std": 0.3762083053588867,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.2920094132423401,
+      "rewards/code_quality_reward_func": 0.2750000059604645,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2666667103767395,
+      "step": 34
+    },
+    {
+      "completion_length": 242.0,
+      "epoch": 0.009365801445009366,
+      "grad_norm": 5.15625,
+      "kl": 0.0007896397146396339,
+      "learning_rate": 1.8716577540106954e-06,
+      "loss": 0.0,
+      "reward": 1.6047223806381226,
+      "reward_std": 0.5125775337219238,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.3463890254497528,
+      "rewards/code_quality_reward_func": 0.30000001192092896,
+      "rewards/reasoning_quality_reward_func": 0.36666667461395264,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 35
+    },
+    {
+      "completion_length": 309.25,
+      "epoch": 0.009633395772009633,
+      "grad_norm": 4.1875,
+      "kl": 0.0007134590996429324,
+      "learning_rate": 1.9251336898395724e-06,
+      "loss": 0.0,
+      "reward": 1.787470817565918,
+      "reward_std": 0.6907145380973816,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.43747079372406006,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 36
+    },
+    {
+      "completion_length": 224.25,
+      "epoch": 0.009900990099009901,
+      "grad_norm": 5.96875,
+      "kl": 0.000888791517354548,
+      "learning_rate": 1.9786096256684497e-06,
+      "loss": 0.0,
+      "reward": 1.5097030401229858,
+      "reward_std": 0.4457584023475647,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.2513695955276489,
+      "rewards/code_quality_reward_func": 0.2916666865348816,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 37
+    },
+    {
+      "completion_length": 264.66668701171875,
+      "epoch": 0.010168584426010168,
+      "grad_norm": 4.125,
+      "kl": 0.0007836823933757842,
+      "learning_rate": 2.0320855614973265e-06,
+      "loss": 0.0,
+      "reward": 2.681027889251709,
+      "reward_std": 0.7436959147453308,
+      "rewards/basic_format_reward_func": 0.5,
+      "rewards/code_execution_reward": 1.139361023902893,
+      "rewards/code_quality_reward_func": 0.3500000238418579,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.30000001192092896,
+      "step": 38
+    },
+    {
+      "completion_length": 251.9166717529297,
+      "epoch": 0.010436178753010436,
+      "grad_norm": 6.1875,
+      "kl": 0.001101348432712257,
+      "learning_rate": 2.0855614973262034e-06,
+      "loss": 0.0,
+      "reward": 2.066322088241577,
+      "reward_std": 0.6089464426040649,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.6413220167160034,
+      "rewards/code_quality_reward_func": 0.3083333671092987,
+      "rewards/reasoning_quality_reward_func": 0.3666667342185974,
+      "rewards/xml_structure_reward_func": 0.2916666865348816,
+      "step": 39
+    },
+    {
+      "completion_length": 254.5,
+      "epoch": 0.010703773080010704,
+      "grad_norm": 5.28125,
+      "kl": 0.001115662045776844,
+      "learning_rate": 2.1390374331550802e-06,
+      "loss": 0.0,
+      "reward": 1.8096225261688232,
+      "reward_std": 0.2221776843070984,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.4096224308013916,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 40
+    },
+    {
+      "completion_length": 217.75,
+      "epoch": 0.010971367407010971,
+      "grad_norm": 5.0,
+      "kl": 0.0014358344487845898,
+      "learning_rate": 2.1925133689839575e-06,
+      "loss": 0.0,
+      "reward": 2.3336026668548584,
+      "reward_std": 1.1396851539611816,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 1.008602499961853,
+      "rewards/code_quality_reward_func": 0.2916666865348816,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 41
+    },
+    {
+      "completion_length": 203.83334350585938,
+      "epoch": 0.01123896173401124,
+      "grad_norm": 6.6875,
+      "kl": 0.0014540926786139607,
+      "learning_rate": 2.2459893048128343e-06,
+      "loss": 0.0,
+      "reward": 1.6667674779891968,
+      "reward_std": 0.40645909309387207,
+      "rewards/basic_format_reward_func": 0.2916666865348816,
+      "rewards/code_execution_reward": 0.4167674481868744,
+      "rewards/code_quality_reward_func": 0.3333333432674408,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.24166667461395264,
+      "step": 42
+    },
+    {
+      "completion_length": 246.08334350585938,
+      "epoch": 0.011506556061011506,
+      "grad_norm": 6.0625,
+      "kl": 0.0013111111475154757,
+      "learning_rate": 2.2994652406417116e-06,
+      "loss": 0.0,
+      "reward": 1.6583411693572998,
+      "reward_std": 0.8617290258407593,
+      "rewards/basic_format_reward_func": 0.2916666865348816,
+      "rewards/code_execution_reward": 0.44167447090148926,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.36666667461395264,
+      "rewards/xml_structure_reward_func": 0.23333334922790527,
+      "step": 43
+    },
+    {
+      "completion_length": 267.5833435058594,
+      "epoch": 0.011774150388011774,
+      "grad_norm": 5.53125,
+      "kl": 0.00108270940836519,
+      "learning_rate": 2.3529411764705885e-06,
+      "loss": 0.0,
+      "reward": 1.7920515537261963,
+      "reward_std": 0.5655256509780884,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.46705159544944763,
+      "rewards/code_quality_reward_func": 0.38333338499069214,
+      "rewards/reasoning_quality_reward_func": 0.375,
+      "rewards/xml_structure_reward_func": 0.23333333432674408,
+      "step": 44
+    },
+    {
+      "completion_length": 259.8333435058594,
+      "epoch": 0.012041744715012041,
+      "grad_norm": 5.46875,
+      "kl": 0.0007468253606930375,
+      "learning_rate": 2.4064171122994653e-06,
+      "loss": 0.0,
+      "reward": 1.5811223983764648,
+      "reward_std": 0.5154642462730408,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.3477890193462372,
+      "rewards/code_quality_reward_func": 0.2916666865348816,
+      "rewards/reasoning_quality_reward_func": 0.36666667461395264,
+      "rewards/xml_structure_reward_func": 0.24166667461395264,
+      "step": 45
+    },
+    {
+      "completion_length": 322.3333435058594,
+      "epoch": 0.01230933904201231,
+      "grad_norm": 4.53125,
+      "kl": 0.0008717922028154135,
+      "learning_rate": 2.4598930481283426e-06,
+      "loss": 0.0,
+      "reward": 1.7504132986068726,
+      "reward_std": 0.7635027766227722,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.4837464690208435,
+      "rewards/code_quality_reward_func": 0.27500003576278687,
+      "rewards/reasoning_quality_reward_func": 0.36666667461395264,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 46
+    },
+    {
+      "completion_length": 261.5,
+      "epoch": 0.012576933369012578,
+      "grad_norm": 5.25,
+      "kl": 0.0009470974327996373,
+      "learning_rate": 2.5133689839572194e-06,
+      "loss": 0.0,
+      "reward": 1.817216396331787,
+      "reward_std": 0.6923457980155945,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.4338831305503845,
+      "rewards/code_quality_reward_func": 0.3083333373069763,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 47
+    },
+    {
+      "completion_length": 285.0833435058594,
+      "epoch": 0.012844527696012844,
+      "grad_norm": 4.71875,
+      "kl": 0.0011376328766345978,
+      "learning_rate": 2.5668449197860967e-06,
+      "loss": 0.0,
+      "reward": 1.4470218420028687,
+      "reward_std": 0.49360841512680054,
+      "rewards/basic_format_reward_func": 0.25,
+      "rewards/code_execution_reward": 0.30535510182380676,
+      "rewards/code_quality_reward_func": 0.27500003576278687,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.21666666865348816,
+      "step": 48
+    },
+    {
+      "completion_length": 282.75,
+      "epoch": 0.013112122023013113,
+      "grad_norm": 7.4375,
+      "kl": 0.0007720650173723698,
+      "learning_rate": 2.6203208556149735e-06,
+      "loss": 0.0,
+      "reward": 1.6602933406829834,
+      "reward_std": 0.4326972961425781,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.3436264991760254,
+      "rewards/code_quality_reward_func": 0.3083333671092987,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.25,
+      "step": 49
+    },
+    {
+      "completion_length": 258.5833435058594,
+      "epoch": 0.01337971635001338,
+      "grad_norm": 4.96875,
+      "kl": 0.0010720042046159506,
+      "learning_rate": 2.673796791443851e-06,
+      "loss": 0.0,
+      "reward": 1.5806025266647339,
+      "reward_std": 0.44038090109825134,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.29726916551589966,
+      "rewards/code_quality_reward_func": 0.23333334922790527,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2750000059604645,
+      "step": 50
+    },
+    {
+      "completion_length": 299.5,
+      "epoch": 0.013647310677013648,
+      "grad_norm": 4.84375,
+      "kl": 0.0012970238458365202,
+      "learning_rate": 2.7272727272727272e-06,
+      "loss": 0.0,
+      "reward": 2.1270318031311035,
+      "reward_std": 0.6962482929229736,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.7186983823776245,
+      "rewards/code_quality_reward_func": 0.32499998807907104,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2666666805744171,
+      "step": 51
+    },
+    {
+      "completion_length": 331.16668701171875,
+      "epoch": 0.013914905004013914,
+      "grad_norm": 3.578125,
+      "kl": 0.0008269142708741128,
+      "learning_rate": 2.7807486631016045e-06,
+      "loss": 0.0,
+      "reward": 1.9148128032684326,
+      "reward_std": 1.0099449157714844,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.5898126363754272,
+      "rewards/code_quality_reward_func": 0.30000001192092896,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 52
+    },
+    {
+      "completion_length": 253.25,
+      "epoch": 0.014182499331014183,
+      "grad_norm": 5.0625,
+      "kl": 0.001110364100895822,
+      "learning_rate": 2.8342245989304818e-06,
+      "loss": 0.0,
+      "reward": 1.8336787223815918,
+      "reward_std": 0.3221331238746643,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.3503453731536865,
+      "rewards/code_quality_reward_func": 0.3333333432674408,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2916666865348816,
+      "step": 53
+    },
+    {
+      "completion_length": 300.3333435058594,
+      "epoch": 0.014450093658014451,
+      "grad_norm": 5.0625,
+      "kl": 0.0013840228784829378,
+      "learning_rate": 2.8877005347593586e-06,
+      "loss": 0.0,
+      "reward": 1.5333333015441895,
+      "reward_std": 0.46698129177093506,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.24166671931743622,
+      "rewards/code_quality_reward_func": 0.31666669249534607,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 54
+    },
+    {
+      "completion_length": 267.8333435058594,
+      "epoch": 0.014717687985014717,
+      "grad_norm": 4.5625,
+      "kl": 0.0013650886248797178,
+      "learning_rate": 2.9411764705882355e-06,
+      "loss": 0.0,
+      "reward": 2.046318531036377,
+      "reward_std": 0.31367409229278564,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.5546518564224243,
+      "rewards/code_quality_reward_func": 0.3499999940395355,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 55
+    },
+    {
+      "completion_length": 323.0,
+      "epoch": 0.014985282312014986,
+      "grad_norm": 4.5625,
+      "kl": 0.0014160439604893327,
+      "learning_rate": 2.9946524064171123e-06,
+      "loss": 0.0,
+      "reward": 2.1313090324401855,
+      "reward_std": 0.5010348558425903,
+      "rewards/basic_format_reward_func": 0.5,
+      "rewards/code_execution_reward": 0.5813089609146118,
+      "rewards/code_quality_reward_func": 0.3499999940395355,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.30000001192092896,
+      "step": 56
+    },
+    {
+      "completion_length": 264.5,
+      "epoch": 0.015252876639015252,
+      "grad_norm": 4.21875,
+      "kl": 0.0008944954606704414,
+      "learning_rate": 3.0481283422459896e-06,
+      "loss": 0.0,
+      "reward": 1.4083333015441895,
+      "reward_std": 0.21618162095546722,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.05000000447034836,
+      "rewards/code_quality_reward_func": 0.32499998807907104,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 57
+    },
+    {
+      "completion_length": 301.3333435058594,
+      "epoch": 0.01552047096601552,
+      "grad_norm": 5.09375,
+      "kl": 0.0016529497224837542,
+      "learning_rate": 3.101604278074867e-06,
+      "loss": 0.0,
+      "reward": 1.6092885732650757,
+      "reward_std": 0.49159228801727295,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.35928845405578613,
+      "rewards/code_quality_reward_func": 0.3083333671092987,
+      "rewards/reasoning_quality_reward_func": 0.36666667461395264,
+      "rewards/xml_structure_reward_func": 0.24166667461395264,
+      "step": 58
+    },
+    {
+      "completion_length": 236.83334350585938,
+      "epoch": 0.01578806529301579,
+      "grad_norm": 5.6875,
+      "kl": 0.0012853683438152075,
+      "learning_rate": 3.1550802139037433e-06,
+      "loss": 0.0,
+      "reward": 1.714489221572876,
+      "reward_std": 0.7896954417228699,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.40615588426589966,
+      "rewards/code_quality_reward_func": 0.2666666805744171,
+      "rewards/reasoning_quality_reward_func": 0.36666667461395264,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 59
+    },
+    {
+      "completion_length": 285.5833435058594,
+      "epoch": 0.016055659620016056,
+      "grad_norm": 5.0,
+      "kl": 0.0012975976569578052,
+      "learning_rate": 3.2085561497326205e-06,
+      "loss": 0.0,
+      "reward": 2.677506923675537,
+      "reward_std": 0.8623813390731812,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 1.244173526763916,
+      "rewards/code_quality_reward_func": 0.30000001192092896,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 60
+    },
+    {
+      "completion_length": 298.5833435058594,
+      "epoch": 0.016323253947016322,
+      "grad_norm": 4.5625,
+      "kl": 0.0014258928131312132,
+      "learning_rate": 3.262032085561498e-06,
+      "loss": 0.0,
+      "reward": 1.9583333730697632,
+      "reward_std": 0.6213881969451904,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.5333333015441895,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 61
+    },
+    {
+      "completion_length": 360.25,
+      "epoch": 0.016590848274016592,
+      "grad_norm": 4.0,
+      "kl": 0.0012111273827031255,
+      "learning_rate": 3.3155080213903747e-06,
+      "loss": 0.0,
+      "reward": 1.8708466291427612,
+      "reward_std": 0.27652737498283386,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.3958466947078705,
+      "rewards/code_quality_reward_func": 0.3583333194255829,
+      "rewards/reasoning_quality_reward_func": 0.375,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 62
+    },
+    {
+      "completion_length": 282.5833435058594,
+      "epoch": 0.01685844260101686,
+      "grad_norm": 5.21875,
+      "kl": 0.0017485963180661201,
+      "learning_rate": 3.368983957219252e-06,
+      "loss": 0.0,
+      "reward": 1.4168031215667725,
+      "reward_std": 0.46065354347229004,
+      "rewards/basic_format_reward_func": 0.2916666865348816,
+      "rewards/code_execution_reward": 0.24180302023887634,
+      "rewards/code_quality_reward_func": 0.27500003576278687,
+      "rewards/reasoning_quality_reward_func": 0.36666667461395264,
+      "rewards/xml_structure_reward_func": 0.24166667461395264,
+      "step": 63
+    },
+    {
+      "completion_length": 213.58334350585938,
+      "epoch": 0.017126036928017126,
+      "grad_norm": 5.0625,
+      "kl": 0.0008941316627897322,
+      "learning_rate": 3.4224598930481284e-06,
+      "loss": 0.0,
+      "reward": 2.941666603088379,
+      "reward_std": 1.1783628463745117,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 1.5500000715255737,
+      "rewards/code_quality_reward_func": 0.3499999940395355,
+      "rewards/reasoning_quality_reward_func": 0.3499999940395355,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 64
+    },
+    {
+      "completion_length": 299.5833435058594,
+      "epoch": 0.017393631255017392,
+      "grad_norm": 5.3125,
+      "kl": 0.001938142697326839,
+      "learning_rate": 3.4759358288770056e-06,
+      "loss": 0.0,
+      "reward": 1.4340009689331055,
+      "reward_std": 0.5084213018417358,
+      "rewards/basic_format_reward_func": 0.2916666865348816,
+      "rewards/code_execution_reward": 0.2840009927749634,
+      "rewards/code_quality_reward_func": 0.24166667461395264,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.21666668355464935,
+      "step": 65
+    },
+    {
+      "completion_length": 308.66668701171875,
+      "epoch": 0.017661225582017662,
+      "grad_norm": 5.0,
+      "kl": 0.0014387283008545637,
+      "learning_rate": 3.529411764705883e-06,
+      "loss": 0.0,
+      "reward": 1.7985808849334717,
+      "reward_std": 0.411335825920105,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.3735809028148651,
+      "rewards/code_quality_reward_func": 0.3333333432674408,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 66
+    },
+    {
+      "completion_length": 233.75,
+      "epoch": 0.01792881990901793,
+      "grad_norm": 5.46875,
+      "kl": 0.0016164870467036963,
+      "learning_rate": 3.5828877005347597e-06,
+      "loss": 0.0,
+      "reward": 1.8757007122039795,
+      "reward_std": 0.7557878494262695,
+      "rewards/basic_format_reward_func": 0.3333333432674408,
+      "rewards/code_execution_reward": 0.6507007479667664,
+      "rewards/code_quality_reward_func": 0.27500003576278687,
+      "rewards/reasoning_quality_reward_func": 0.375,
+      "rewards/xml_structure_reward_func": 0.24166667461395264,
+      "step": 67
+    },
+    {
+      "completion_length": 327.0,
+      "epoch": 0.018196414236018196,
+      "grad_norm": 5.0625,
+      "kl": 0.002256980864331126,
+      "learning_rate": 3.6363636363636366e-06,
+      "loss": 0.0,
+      "reward": 1.734962821006775,
+      "reward_std": 0.3988541066646576,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.2682960629463196,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 68
+    },
+    {
+      "completion_length": 348.91668701171875,
+      "epoch": 0.018464008563018466,
+      "grad_norm": 4.0,
+      "kl": 0.0022144902031868696,
+      "learning_rate": 3.6898395721925134e-06,
+      "loss": 0.0,
+      "reward": 1.6737807989120483,
+      "reward_std": 0.551749587059021,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.3404473066329956,
+      "rewards/code_quality_reward_func": 0.2916666865348816,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2666666805744171,
+      "step": 69
+    },
+    {
+      "completion_length": 211.58334350585938,
+      "epoch": 0.018731602890018732,
+      "grad_norm": 5.5,
+      "kl": 0.0028423962648957968,
+      "learning_rate": 3.7433155080213907e-06,
+      "loss": 0.0,
+      "reward": 2.084716558456421,
+      "reward_std": 0.7005678415298462,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.6347164511680603,
+      "rewards/code_quality_reward_func": 0.3083333373069763,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.2916666865348816,
+      "step": 70
+    },
+    {
+      "completion_length": 283.8333435058594,
+      "epoch": 0.018999197217019,
+      "grad_norm": 4.53125,
+      "kl": 0.0025364342145621777,
+      "learning_rate": 3.796791443850268e-06,
+      "loss": 0.0,
+      "reward": 1.9688270092010498,
+      "reward_std": 1.0129854679107666,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.6854934692382812,
+      "rewards/code_quality_reward_func": 0.25833338499069214,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 71
+    },
+    {
+      "completion_length": 273.16668701171875,
+      "epoch": 0.019266791544019266,
+      "grad_norm": 6.21875,
+      "kl": 0.001806919346563518,
+      "learning_rate": 3.850267379679145e-06,
+      "loss": 0.0,
+      "reward": 2.2200722694396973,
+      "reward_std": 0.6418557167053223,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.7534055709838867,
+      "rewards/code_quality_reward_func": 0.3333333432674408,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.2916666865348816,
+      "step": 72
+    },
+    {
+      "completion_length": 253.5,
+      "epoch": 0.019534385871019536,
+      "grad_norm": 6.1875,
+      "kl": 0.002755642170086503,
+      "learning_rate": 3.903743315508022e-06,
+      "loss": 0.0,
+      "reward": 1.7943209409713745,
+      "reward_std": 0.4601937234401703,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.47765427827835083,
+      "rewards/code_quality_reward_func": 0.30000001192092896,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 73
+    },
+    {
+      "completion_length": 222.9166717529297,
+      "epoch": 0.019801980198019802,
+      "grad_norm": 6.03125,
+      "kl": 0.002581920474767685,
+      "learning_rate": 3.957219251336899e-06,
+      "loss": 0.0,
+      "reward": 1.7171714305877686,
+      "reward_std": 0.35186177492141724,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.3255046010017395,
+      "rewards/code_quality_reward_func": 0.2666667103767395,
+      "rewards/reasoning_quality_reward_func": 0.375,
+      "rewards/xml_structure_reward_func": 0.2916666865348816,
+      "step": 74
+    },
+    {
+      "completion_length": 342.75,
+      "epoch": 0.02006957452502007,
+      "grad_norm": 4.6875,
+      "kl": 0.0017477937508374453,
+      "learning_rate": 4.010695187165775e-06,
+      "loss": 0.0,
+      "reward": 1.7918099164962769,
+      "reward_std": 0.33391231298446655,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.29180991649627686,
+      "rewards/code_quality_reward_func": 0.3499999940395355,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2916666865348816,
+      "step": 75
+    },
+    {
+      "completion_length": 277.5,
+      "epoch": 0.020337168852020335,
+      "grad_norm": 5.46875,
+      "kl": 0.002431904897093773,
+      "learning_rate": 4.064171122994653e-06,
+      "loss": 0.0,
+      "reward": 1.6002721786499023,
+      "reward_std": 0.4187708795070648,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.27527207136154175,
+      "rewards/code_quality_reward_func": 0.3166666626930237,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.2500000298023224,
+      "step": 76
+    },
+    {
+      "completion_length": 257.5833435058594,
+      "epoch": 0.020604763179020606,
+      "grad_norm": 5.21875,
+      "kl": 0.0018626938108354807,
+      "learning_rate": 4.11764705882353e-06,
+      "loss": 0.0,
+      "reward": 1.8927496671676636,
+      "reward_std": 0.4733630120754242,
+      "rewards/basic_format_reward_func": 0.5,
+      "rewards/code_execution_reward": 0.4094163775444031,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.3583333492279053,
+      "rewards/xml_structure_reward_func": 0.30000001192092896,
+      "step": 77
+    },
+    {
+      "completion_length": 258.25,
+      "epoch": 0.020872357506020872,
+      "grad_norm": 5.0,
+      "kl": 0.00294861220754683,
+      "learning_rate": 4.171122994652407e-06,
+      "loss": 0.0,
+      "reward": 2.0084242820739746,
+      "reward_std": 0.6261391639709473,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.5584243535995483,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 78
+    },
+    {
+      "completion_length": 231.58334350585938,
+      "epoch": 0.02113995183302114,
+      "grad_norm": 4.40625,
+      "kl": 0.002004222711548209,
+      "learning_rate": 4.224598930481284e-06,
+      "loss": 0.0,
+      "reward": 2.4067962169647217,
+      "reward_std": 0.8294322490692139,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.9734629392623901,
+      "rewards/code_quality_reward_func": 0.3166666626930237,
+      "rewards/reasoning_quality_reward_func": 0.375,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 79
+    },
+    {
+      "completion_length": 291.25,
+      "epoch": 0.02140754616002141,
+      "grad_norm": 5.40625,
+      "kl": 0.004346501547843218,
+      "learning_rate": 4.2780748663101604e-06,
+      "loss": 0.0,
+      "reward": 1.777273178100586,
+      "reward_std": 0.38957223296165466,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.3606064021587372,
+      "rewards/code_quality_reward_func": 0.34166666865348816,
+      "rewards/reasoning_quality_reward_func": 0.375,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 80
+    },
+    {
+      "completion_length": 234.5,
+      "epoch": 0.021675140487021675,
+      "grad_norm": 6.3125,
+      "kl": 0.0035126435104757547,
+      "learning_rate": 4.331550802139038e-06,
+      "loss": 0.0,
+      "reward": 1.897292971611023,
+      "reward_std": 0.42207369208335876,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.4889596402645111,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.3916666507720947,
+      "rewards/xml_structure_reward_func": 0.2750000059604645,
+      "step": 81
+    },
+    {
+      "completion_length": 228.9166717529297,
+      "epoch": 0.021942734814021942,
+      "grad_norm": 6.28125,
+      "kl": 0.002383569488301873,
+      "learning_rate": 4.385026737967915e-06,
+      "loss": 0.0,
+      "reward": 1.6903694868087769,
+      "reward_std": 0.436374306678772,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.3487027585506439,
+      "rewards/code_quality_reward_func": 0.31666669249534607,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 82
+    },
+    {
+      "completion_length": 297.0,
+      "epoch": 0.02221032914102221,
+      "grad_norm": 5.28125,
+      "kl": 0.0043878196738660336,
+      "learning_rate": 4.438502673796792e-06,
+      "loss": 0.0,
+      "reward": 1.7573189735412598,
+      "reward_std": 0.3008286654949188,
+      "rewards/basic_format_reward_func": 0.5,
+      "rewards/code_execution_reward": 0.2573189437389374,
+      "rewards/code_quality_reward_func": 0.30000001192092896,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.30000001192092896,
+      "step": 83
+    },
+    {
+      "completion_length": 242.75,
+      "epoch": 0.02247792346802248,
+      "grad_norm": 7.75,
+      "kl": 0.0079798623919487,
+      "learning_rate": 4.491978609625669e-06,
+      "loss": 0.0,
+      "reward": 1.7856076955795288,
+      "reward_std": 0.4495140016078949,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.4022744297981262,
+      "rewards/code_quality_reward_func": 0.31666669249534607,
+      "rewards/reasoning_quality_reward_func": 0.375,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 84
+    },
+    {
+      "completion_length": 246.75,
+      "epoch": 0.022745517795022745,
+      "grad_norm": 5.03125,
+      "kl": 0.0022815698757767677,
+      "learning_rate": 4.5454545454545455e-06,
+      "loss": 0.0,
+      "reward": 1.8084133863449097,
+      "reward_std": 0.18776123225688934,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.31674671173095703,
+      "rewards/code_quality_reward_func": 0.34166669845581055,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2916666865348816,
+      "step": 85
+    },
+    {
+      "completion_length": 305.5,
+      "epoch": 0.023013112122023012,
+      "grad_norm": 4.21875,
+      "kl": 0.003150416072458029,
+      "learning_rate": 4.598930481283423e-06,
+      "loss": 0.0,
+      "reward": 1.6666734218597412,
+      "reward_std": 0.3055960536003113,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.2250065803527832,
+      "rewards/code_quality_reward_func": 0.3083333373069763,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 86
+    },
+    {
+      "completion_length": 185.1666717529297,
+      "epoch": 0.023280706449023282,
+      "grad_norm": 5.1875,
+      "kl": 0.005251727066934109,
+      "learning_rate": 4.6524064171123e-06,
+      "loss": 0.0,
+      "reward": 2.542512893676758,
+      "reward_std": 0.6368582248687744,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 1.1925126314163208,
+      "rewards/code_quality_reward_func": 0.3166666626930237,
+      "rewards/reasoning_quality_reward_func": 0.34166669845581055,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 87
+    },
+    {
+      "completion_length": 298.8333435058594,
+      "epoch": 0.02354830077602355,
+      "grad_norm": 5.21875,
+      "kl": 0.006109694484621286,
+      "learning_rate": 4.705882352941177e-06,
+      "loss": 0.0,
+      "reward": 1.9166667461395264,
+      "reward_std": 0.6324446201324463,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.5333333015441895,
+      "rewards/code_quality_reward_func": 0.30000001192092896,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.2666667103767395,
+      "step": 88
+    },
+    {
+      "completion_length": 216.25,
+      "epoch": 0.023815895103023815,
+      "grad_norm": 5.03125,
+      "kl": 0.0036100514698773623,
+      "learning_rate": 4.759358288770054e-06,
+      "loss": 0.0,
+      "reward": 2.4646191596984863,
+      "reward_std": 0.5950441360473633,
+      "rewards/basic_format_reward_func": 0.5,
+      "rewards/code_execution_reward": 0.9646190404891968,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.375,
+      "rewards/xml_structure_reward_func": 0.30000001192092896,
+      "step": 89
+    },
+    {
+      "completion_length": 212.1666717529297,
+      "epoch": 0.024083489430024082,
+      "grad_norm": 5.625,
+      "kl": 0.004241304472088814,
+      "learning_rate": 4.812834224598931e-06,
+      "loss": 0.0,
+      "reward": 1.717494249343872,
+      "reward_std": 0.3651425540447235,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.3341608941555023,
+      "rewards/code_quality_reward_func": 0.3083333373069763,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.27500003576278687,
+      "step": 90
+    },
+    {
+      "completion_length": 253.1666717529297,
+      "epoch": 0.024351083757024352,
+      "grad_norm": 6.21875,
+      "kl": 0.01768430508673191,
+      "learning_rate": 4.866310160427808e-06,
+      "loss": 0.0,
+      "reward": 1.7001416683197021,
+      "reward_std": 0.21167293190956116,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.24180833995342255,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.2916666865348816,
+      "step": 91
+    },
+    {
+      "completion_length": 268.16668701171875,
+      "epoch": 0.02461867808402462,
+      "grad_norm": 4.46875,
+      "kl": 0.0029091592878103256,
+      "learning_rate": 4.919786096256685e-06,
+      "loss": 0.0,
+      "reward": 2.156282424926758,
+      "reward_std": 0.7013486623764038,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.7146157026290894,
+      "rewards/code_quality_reward_func": 0.31666669249534607,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 92
+    },
+    {
+      "completion_length": 213.9166717529297,
+      "epoch": 0.024886272411024885,
+      "grad_norm": 5.96875,
+      "kl": 0.0043329414911568165,
+      "learning_rate": 4.973262032085562e-06,
+      "loss": 0.0,
+      "reward": 1.6091450452804565,
+      "reward_std": 0.4102088212966919,
+      "rewards/basic_format_reward_func": 0.375,
+      "rewards/code_execution_reward": 0.30081164836883545,
+      "rewards/code_quality_reward_func": 0.28333336114883423,
+      "rewards/reasoning_quality_reward_func": 0.3916666507720947,
+      "rewards/xml_structure_reward_func": 0.25833335518836975,
+      "step": 93
+    },
+    {
+      "completion_length": 297.3333435058594,
+      "epoch": 0.025153866738025155,
+      "grad_norm": 5.21875,
+      "kl": 0.0038607739843428135,
+      "learning_rate": 5.026737967914439e-06,
+      "loss": 0.0,
+      "reward": 1.887909173965454,
+      "reward_std": 0.30988383293151855,
+      "rewards/basic_format_reward_func": 0.5,
+      "rewards/code_execution_reward": 0.35457563400268555,
+      "rewards/code_quality_reward_func": 0.3333333432674408,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.30000001192092896,
+      "step": 94
+    },
+    {
+      "completion_length": 237.08334350585938,
+      "epoch": 0.025421461065025422,
+      "grad_norm": 6.09375,
+      "kl": 0.0038710730150341988,
+      "learning_rate": 5.0802139037433165e-06,
+      "loss": 0.0,
+      "reward": 1.8167171478271484,
+      "reward_std": 0.3526454567909241,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.41671714186668396,
+      "rewards/code_quality_reward_func": 0.31666669249534607,
+      "rewards/reasoning_quality_reward_func": 0.3916667103767395,
+      "rewards/xml_structure_reward_func": 0.2750000059604645,
+      "step": 95
+    },
+    {
+      "completion_length": 269.16668701171875,
+      "epoch": 0.02568905539202569,
+      "grad_norm": 5.3125,
+      "kl": 0.0043390365317463875,
+      "learning_rate": 5.133689839572193e-06,
+      "loss": 0.0,
+      "reward": 1.9732608795166016,
+      "reward_std": 0.4741336703300476,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.4982609748840332,
+      "rewards/code_quality_reward_func": 0.3333333432674408,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 96
+    },
+    {
+      "completion_length": 253.1666717529297,
+      "epoch": 0.025956649719025955,
+      "grad_norm": 6.03125,
+      "kl": 0.006835515145212412,
+      "learning_rate": 5.187165775401069e-06,
+      "loss": 0.0,
+      "reward": 1.851419448852539,
+      "reward_std": 0.38760122656822205,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.40141940116882324,
+      "rewards/code_quality_reward_func": 0.32500001788139343,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 97
+    },
+    {
+      "completion_length": 287.66668701171875,
+      "epoch": 0.026224244046026225,
+      "grad_norm": 4.4375,
+      "kl": 0.004592955112457275,
+      "learning_rate": 5.240641711229947e-06,
+      "loss": 0.0,
+      "reward": 1.9823663234710693,
+      "reward_std": 0.15290634334087372,
+      "rewards/basic_format_reward_func": 0.5,
+      "rewards/code_execution_reward": 0.4323663115501404,
+      "rewards/code_quality_reward_func": 0.3499999940395355,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.30000001192092896,
+      "step": 98
+    },
+    {
+      "completion_length": 291.0,
+      "epoch": 0.026491838373026492,
+      "grad_norm": 4.46875,
+      "kl": 0.004338396713137627,
+      "learning_rate": 5.294117647058824e-06,
+      "loss": 0.0,
+      "reward": 1.9069900512695312,
+      "reward_std": 0.4237919747829437,
+      "rewards/basic_format_reward_func": 0.4166666865348816,
+      "rewards/code_execution_reward": 0.48199015855789185,
+      "rewards/code_quality_reward_func": 0.34166666865348816,
+      "rewards/reasoning_quality_reward_func": 0.38333338499069214,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 99
+    },
+    {
+      "completion_length": 336.16668701171875,
+      "epoch": 0.02675943270002676,
+      "grad_norm": 4.59375,
+      "kl": 0.004567224998027086,
+      "learning_rate": 5.347593582887702e-06,
+      "loss": 0.0,
+      "reward": 1.897521734237671,
+      "reward_std": 0.35646283626556396,
+      "rewards/basic_format_reward_func": 0.4583333432674408,
+      "rewards/code_execution_reward": 0.46418827772140503,
+      "rewards/code_quality_reward_func": 0.2916666865348816,
+      "rewards/reasoning_quality_reward_func": 0.40000003576278687,
+      "rewards/xml_structure_reward_func": 0.28333336114883423,
+      "step": 100
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 3737,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "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": 1,
+  "trial_name": null,
+  "trial_params": null
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cabc81062865544af01bbdc9d45a6f3370b497f42952f0b634459bb2ea96fca0
+size 5624

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c1d2d2c665a761dca1a6e793aa0fcc1f9a8bb4b71c0ecf6d7ce3ca7e21325f5
+size 1976374202

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1200/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a181f049b8b4c72b9860feec9f2ab6994db6a7224a6e62e162d8ca1e310b01e1
+size 988097824

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1200/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73c775c8033ac05ae05bf9f4be96ca11345623bbd6e8446ba22228e55af09e06
+size 1976374202

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1300/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77de3a3375c53333ea216a0fa9543603557da31efb975bdbcfa6c9fdc51a91c1
+size 988097824

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1300/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fc00ae97486588443c61b4f67a4b772c92238025bc6fe377e4428fcf17d1dda
+size 1976374202

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/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
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "./outputs/Qwen-0.5B-GRPO-Code-vllm-base/checkpoint-800",
+  "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": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/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": "4.48.2"
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:440bade28ce8f58bec094cac3fc7ce45c546182369d785dbd81dc661ab012a68
+size 14448

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f68a37892a1b445d21bb35cc10bf7a058a6f9ec8c363f5ed156ff4f49d90fb6
+size 14512

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8754a5f1cbd85f66d1ea8b9604fdbd58dbfd1eb629a110e719621dd41c549896
+size 1064

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "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": "<|im_end|>"
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63a2951d5edfa5cc0a2346ef872f8c77a2920274cfc3b503b04e3799104dee80
+size 11422060

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/tokenizer_config.json

@@ -0,0 +1,212 @@
+{
+  "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": {},
+  "max_length": null,
+  "model_max_length": 131072,
+  "pad_to_multiple_of": null,
+  "pad_token": "<|im_end|>",
+  "pad_token_type_id": 0,
+  "padding_side": "left",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1600/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1a7fd0df466b8805938262d9ba986467f52c4e050d689524d6c7e1032bae220
+size 5624

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1700/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f950502b07530b259c47a9b9ce3260c30fd52047d833d179d2ec4a13e5585dd5
+size 988097824

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-1900/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13884a01e434329b3500b37bf78d852b5bd3df1c56a7eacdf270c2779a52ee64
+size 1976374202

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-200/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a471a4311eb3157e819e65ece392245e98c77162ec74c811ffc233c8d648f2f2
+size 988097824

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-2000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:265e4ed1e3e20f31081e57cde3d0338204cf46072cd3882236ff01720d78a066
+size 988097824

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-2100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abc84d8812032dfb6aa5357e74c47048214de0275a9992ac244e0e8ccb04cc68
+size 1976374202

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99f095b72ba0a54812ccc7fa56c7d23aaea57ec0094fa391ffcb1f205a97c76c
+size 988097824

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-500/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a55814bc8055c887a97bd0847ae903851a377d015db4ae6afa6b2a4308fef8b7
+size 14448

Qwen-0.5B-GRPO-Code-vllm-modified-v1/checkpoint-500/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f68a37892a1b445d21bb35cc10bf7a058a6f9ec8c363f5ed156ff4f49d90fb6
+size 14512