model commit

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

added_tokens.json

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

config.json

@@ -0,0 +1,28 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 896,
+  "initializer_range": 0.02,
+  "intermediate_size": 4864,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "qwen2",
+  "num_attention_heads": 14,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.0",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.51.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbdd3fdd56b8102e9c73126018ba93d9c6e2255f9931d0c63d739ae06bedbacc
+size 988097824

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
+size 11422063

tokenizer_config.json

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

trainer_state.json

@@ -0,0 +1,1634 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio": 0.003959985822439194,
+      "completion_length": 54.13888931274414,
+      "epoch": 0.01,
+      "grad_norm": 784750.1992774517,
+      "kl": 785907.2,
+      "learning_rate": 4e-07,
+      "loss": 1571.24,
+      "reward": 0.09333334118127823,
+      "reward_std": 0.15150094280640283,
+      "rewards/judge_tool_use": 0.09236111243565877,
+      "rewards/judge_tool_use/std": 0.22503272195657095,
+      "rewards/verify_correctness": 0.09722222325702508,
+      "rewards/verify_correctness/std": 0.27655620872974396,
+      "step": 5
+    },
+    {
+      "clip_ratio": 0.007778534106910229,
+      "completion_length": 70.86458587646484,
+      "epoch": 0.02,
+      "grad_norm": 7994810.896724847,
+      "kl": 36229.2,
+      "learning_rate": 9e-07,
+      "loss": 72.2516,
+      "reward": 0.1066666692495346,
+      "reward_std": 0.19039028137922287,
+      "rewards/judge_tool_use": 0.12291667237877846,
+      "rewards/judge_tool_use/std": 0.2716349810361862,
+      "rewards/verify_correctness": 0.0416666679084301,
+      "rewards/verify_correctness/std": 0.20194092392921448,
+      "step": 10
+    },
+    {
+      "clip_ratio": 0.006301742233335972,
+      "completion_length": 75.5069465637207,
+      "epoch": 0.03,
+      "grad_norm": 10255.043920889624,
+      "kl": 26379.2,
+      "learning_rate": 1e-06,
+      "loss": 52.6755,
+      "reward": 0.06888888776302338,
+      "reward_std": 0.12580444912115732,
+      "rewards/judge_tool_use": 0.07222222536802292,
+      "rewards/judge_tool_use/std": 0.19341551760832468,
+      "rewards/verify_correctness": 0.055555555348594986,
+      "rewards/verify_correctness/std": 0.21832925577958426,
+      "step": 15
+    },
+    {
+      "clip_ratio": 0.005547903012484312,
+      "completion_length": 63.864585876464844,
+      "epoch": 0.04,
+      "grad_norm": 12547.287212363904,
+      "kl": 438.30462646484375,
+      "learning_rate": 1e-06,
+      "loss": 0.8478,
+      "reward": 0.03999999910593033,
+      "reward_std": 0.06966040655970573,
+      "rewards/judge_tool_use": 0.03437500377185643,
+      "rewards/judge_tool_use/std": 0.10251733846962452,
+      "rewards/verify_correctness": 0.0625,
+      "rewards/verify_correctness/std": 0.16710934042930603,
+      "step": 20
+    },
+    {
+      "clip_ratio": 0.004662872105836868,
+      "completion_length": 89.24305725097656,
+      "epoch": 0.05,
+      "grad_norm": 298314332.7497967,
+      "kl": 204444.275,
+      "learning_rate": 1e-06,
+      "loss": 406.5101,
+      "reward": 0.12250000486771266,
+      "reward_std": 0.172040027876695,
+      "rewards/judge_tool_use": 0.13055556019147238,
+      "rewards/judge_tool_use/std": 0.2499552220106125,
+      "rewards/verify_correctness": 0.09027777860562007,
+      "rewards/verify_correctness/std": 0.22458433111508688,
+      "step": 25
+    },
+    {
+      "clip_ratio": 0.007842596806585788,
+      "completion_length": 125.81250381469727,
+      "epoch": 0.06,
+      "grad_norm": 2920.4418807388934,
+      "kl": 21399.8,
+      "learning_rate": 1e-06,
+      "loss": 42.6862,
+      "reward": 0.16333334147930145,
+      "reward_std": 0.17722879350185394,
+      "rewards/judge_tool_use": 0.17812500149011612,
+      "rewards/judge_tool_use/std": 0.2530943974852562,
+      "rewards/verify_correctness": 0.1041666641831398,
+      "rewards/verify_correctness/std": 0.20520704984664917,
+      "step": 30
+    },
+    {
+      "clip_ratio": 0.006266751885414123,
+      "completion_length": 117.37500254313152,
+      "epoch": 0.07,
+      "grad_norm": 24639.61780550995,
+      "kl": 91.1203125,
+      "learning_rate": 1e-06,
+      "loss": 0.1544,
+      "reward": 0.21361111104488373,
+      "reward_std": 0.2253884275754293,
+      "rewards/judge_tool_use": 0.23055555919806162,
+      "rewards/judge_tool_use/std": 0.33459242184956867,
+      "rewards/verify_correctness": 0.14583333830038706,
+      "rewards/verify_correctness/std": 0.3359044094880422,
+      "step": 35
+    },
+    {
+      "clip_ratio": 0.008563124015927315,
+      "completion_length": 122.10416793823242,
+      "epoch": 0.08,
+      "grad_norm": 220.2399474774772,
+      "kl": 250.11875,
+      "learning_rate": 1e-06,
+      "loss": 0.552,
+      "reward": 0.21416668593883514,
+      "reward_std": 0.21450992673635483,
+      "rewards/judge_tool_use": 0.23125002533197403,
+      "rewards/judge_tool_use/std": 0.30767104029655457,
+      "rewards/verify_correctness": 0.14583333861082792,
+      "rewards/verify_correctness/std": 0.2967151999473572,
+      "step": 40
+    },
+    {
+      "clip_ratio": 0.005935636721551419,
+      "completion_length": 130.95833841959634,
+      "epoch": 0.09,
+      "grad_norm": 5460.174774182503,
+      "kl": 244.15,
+      "learning_rate": 1e-06,
+      "loss": 0.5043,
+      "reward": 0.2500000099341075,
+      "reward_std": 0.23432709276676178,
+      "rewards/judge_tool_use": 0.2743055472771327,
+      "rewards/judge_tool_use/std": 0.3266789714495341,
+      "rewards/verify_correctness": 0.1527777761220932,
+      "rewards/verify_correctness/std": 0.36177847782770794,
+      "step": 45
+    },
+    {
+      "clip_ratio": 0.006385140307247639,
+      "completion_length": 151.70833587646484,
+      "epoch": 0.1,
+      "grad_norm": 30.233107092288996,
+      "kl": 5.925,
+      "learning_rate": 1e-06,
+      "loss": 0.187,
+      "reward": 0.16625000536441803,
+      "reward_std": 0.15779273211956024,
+      "rewards/judge_tool_use": 0.17916666343808174,
+      "rewards/judge_tool_use/std": 0.24784564971923828,
+      "rewards/verify_correctness": 0.11458333395421505,
+      "rewards/verify_correctness/std": 0.2981918603181839,
+      "step": 50
+    },
+    {
+      "clip_ratio": 0.005314220301806926,
+      "completion_length": 133.38194529215494,
+      "epoch": 0.11,
+      "grad_norm": 116.04184304342279,
+      "kl": 5.9447265625,
+      "learning_rate": 1e-06,
+      "loss": 0.0148,
+      "reward": 0.27722224096457165,
+      "reward_std": 0.2598785211642583,
+      "rewards/judge_tool_use": 0.2979166582226753,
+      "rewards/judge_tool_use/std": 0.3200907011826833,
+      "rewards/verify_correctness": 0.1944444477558136,
+      "rewards/verify_correctness/std": 0.37710362672805786,
+      "step": 55
+    },
+    {
+      "clip_ratio": 0.007683201320469379,
+      "completion_length": 126.04166793823242,
+      "epoch": 0.12,
+      "grad_norm": 300.13848407753744,
+      "kl": 2.870703125,
+      "learning_rate": 1e-06,
+      "loss": 0.0265,
+      "reward": 0.22583334147930145,
+      "reward_std": 0.18492110818624496,
+      "rewards/judge_tool_use": 0.2562500014901161,
+      "rewards/judge_tool_use/std": 0.3037812113761902,
+      "rewards/verify_correctness": 0.1041666641831398,
+      "rewards/verify_correctness/std": 0.300648495554924,
+      "step": 60
+    },
+    {
+      "clip_ratio": 0.0056252704933285715,
+      "completion_length": 121.22916666666667,
+      "epoch": 0.13,
+      "grad_norm": 10523.81268433656,
+      "kl": 35.8203125,
+      "learning_rate": 1e-06,
+      "loss": 0.0502,
+      "reward": 0.30444446206092834,
+      "reward_std": 0.20237582921981812,
+      "rewards/judge_tool_use": 0.33541667461395264,
+      "rewards/judge_tool_use/std": 0.3482949535051982,
+      "rewards/verify_correctness": 0.18055555721124014,
+      "rewards/verify_correctness/std": 0.3799951672554016,
+      "step": 65
+    },
+    {
+      "clip_ratio": 0.008523338288068772,
+      "completion_length": 109.78125381469727,
+      "epoch": 0.14,
+      "grad_norm": 306.75431905722485,
+      "kl": 54.10625,
+      "learning_rate": 1e-06,
+      "loss": 0.1497,
+      "reward": 0.37958335876464844,
+      "reward_std": 0.27913033962249756,
+      "rewards/judge_tool_use": 0.39374998211860657,
+      "rewards/judge_tool_use/std": 0.3710695207118988,
+      "rewards/verify_correctness": 0.3229166716337204,
+      "rewards/verify_correctness/std": 0.47240525484085083,
+      "step": 70
+    },
+    {
+      "clip_ratio": 0.006300021335482598,
+      "completion_length": 113.02777862548828,
+      "epoch": 0.15,
+      "grad_norm": 276.105755604053,
+      "kl": 120.347265625,
+      "learning_rate": 1e-06,
+      "loss": 0.2603,
+      "reward": 0.3752777973810832,
+      "reward_std": 0.23248672982056937,
+      "rewards/judge_tool_use": 0.39097222685813904,
+      "rewards/judge_tool_use/std": 0.39233259359995526,
+      "rewards/verify_correctness": 0.3125,
+      "rewards/verify_correctness/std": 0.4657697280248006,
+      "step": 75
+    },
+    {
+      "clip_ratio": 0.008843735605478287,
+      "completion_length": 138.8854217529297,
+      "epoch": 0.16,
+      "grad_norm": 1275.1249001898307,
+      "kl": 7.355078125,
+      "learning_rate": 1e-06,
+      "loss": 0.0298,
+      "reward": 0.28958334028720856,
+      "reward_std": 0.21345646679401398,
+      "rewards/judge_tool_use": 0.3072916716337204,
+      "rewards/judge_tool_use/std": 0.3630993217229843,
+      "rewards/verify_correctness": 0.21875,
+      "rewards/verify_correctness/std": 0.41601888835430145,
+      "step": 80
+    },
+    {
+      "clip_ratio": 0.007743672281503677,
+      "completion_length": 101.95833587646484,
+      "epoch": 0.17,
+      "grad_norm": 890.3990436610322,
+      "kl": 63141.15,
+      "learning_rate": 1e-06,
+      "loss": 125.8673,
+      "reward": 0.5141666730244955,
+      "reward_std": 0.22655925651391348,
+      "rewards/judge_tool_use": 0.5298611124356588,
+      "rewards/judge_tool_use/std": 0.38938658436139423,
+      "rewards/verify_correctness": 0.4513888955116272,
+      "rewards/verify_correctness/std": 0.4737345178922017,
+      "step": 85
+    },
+    {
+      "clip_ratio": 0.009472730197012424,
+      "completion_length": 124.94791793823242,
+      "epoch": 0.18,
+      "grad_norm": 172.8478197001071,
+      "kl": 7.838671875,
+      "learning_rate": 1e-06,
+      "loss": 0.0279,
+      "reward": 0.49916666746139526,
+      "reward_std": 0.17976997047662735,
+      "rewards/judge_tool_use": 0.5250000059604645,
+      "rewards/judge_tool_use/std": 0.3990606963634491,
+      "rewards/verify_correctness": 0.3958333283662796,
+      "rewards/verify_correctness/std": 0.4937349408864975,
+      "step": 90
+    },
+    {
+      "clip_ratio": 0.0054151594638824465,
+      "completion_length": 127.84027862548828,
+      "epoch": 0.19,
+      "grad_norm": 2053.861574982554,
+      "kl": 26665.05,
+      "learning_rate": 1e-06,
+      "loss": 53.3674,
+      "reward": 0.3519444515307744,
+      "reward_std": 0.16984381278355917,
+      "rewards/judge_tool_use": 0.3791666825612386,
+      "rewards/judge_tool_use/std": 0.35421761870384216,
+      "rewards/verify_correctness": 0.24305555721124014,
+      "rewards/verify_correctness/std": 0.4247280756632487,
+      "step": 95
+    },
+    {
+      "clip_ratio": 0.0074031388387084006,
+      "completion_length": 107.06250381469727,
+      "epoch": 0.2,
+      "grad_norm": 79.32245833730725,
+      "kl": 20.08984375,
+      "learning_rate": 1e-06,
+      "loss": 0.049,
+      "reward": 0.39124999940395355,
+      "reward_std": 0.16201016306877136,
+      "rewards/judge_tool_use": 0.41874998807907104,
+      "rewards/judge_tool_use/std": 0.36362142860889435,
+      "rewards/verify_correctness": 0.28125,
+      "rewards/verify_correctness/std": 0.45300620794296265,
+      "step": 100
+    },
+    {
+      "clip_ratio": 0.00740175861865282,
+      "completion_length": 97.12500254313152,
+      "epoch": 0.21,
+      "grad_norm": 31870.333648680495,
+      "kl": 106.084375,
+      "learning_rate": 1e-06,
+      "loss": 0.297,
+      "reward": 0.38583337267239887,
+      "reward_std": 0.20607438186804453,
+      "rewards/judge_tool_use": 0.42152780294418335,
+      "rewards/judge_tool_use/std": 0.36684316396713257,
+      "rewards/verify_correctness": 0.2430555522441864,
+      "rewards/verify_correctness/std": 0.41877804199854535,
+      "step": 105
+    },
+    {
+      "clip_ratio": 0.006064230762422085,
+      "completion_length": 122.32292175292969,
+      "epoch": 0.22,
+      "grad_norm": 12.413314903140305,
+      "kl": 28.44140625,
+      "learning_rate": 1e-06,
+      "loss": 0.0699,
+      "reward": 0.33916667103767395,
+      "reward_std": 0.11534206941723824,
+      "rewards/judge_tool_use": 0.36666668951511383,
+      "rewards/judge_tool_use/std": 0.3902362436056137,
+      "rewards/verify_correctness": 0.2291666641831398,
+      "rewards/verify_correctness/std": 0.41798071563243866,
+      "step": 110
+    },
+    {
+      "clip_ratio": 0.0057834615930914875,
+      "completion_length": 118.7013931274414,
+      "epoch": 0.23,
+      "grad_norm": 10.45791929222894,
+      "kl": 1.521484375,
+      "learning_rate": 1e-06,
+      "loss": 0.025,
+      "reward": 0.3566666742165883,
+      "reward_std": 0.17904378970464072,
+      "rewards/judge_tool_use": 0.39027777314186096,
+      "rewards/judge_tool_use/std": 0.3494054178396861,
+      "rewards/verify_correctness": 0.2222222238779068,
+      "rewards/verify_correctness/std": 0.39503220717112225,
+      "step": 115
+    },
+    {
+      "clip_ratio": 0.008998725935816765,
+      "completion_length": 109.9375,
+      "epoch": 0.24,
+      "grad_norm": 3.4155506219796594,
+      "kl": 0.49375,
+      "learning_rate": 1e-06,
+      "loss": 0.0181,
+      "reward": 0.4558333456516266,
+      "reward_std": 0.1414913684129715,
+      "rewards/judge_tool_use": 0.5072916746139526,
+      "rewards/judge_tool_use/std": 0.38790474832057953,
+      "rewards/verify_correctness": 0.2500000074505806,
+      "rewards/verify_correctness/std": 0.43691860139369965,
+      "step": 120
+    },
+    {
+      "clip_ratio": 0.004882402718067169,
+      "completion_length": 88.28472391764323,
+      "epoch": 0.25,
+      "grad_norm": 6.549383441404032,
+      "kl": 0.5650390625,
+      "learning_rate": 1e-06,
+      "loss": 0.0137,
+      "reward": 0.5822222232818604,
+      "reward_std": 0.2408840556939443,
+      "rewards/judge_tool_use": 0.6062499980131785,
+      "rewards/judge_tool_use/std": 0.3277636071046193,
+      "rewards/verify_correctness": 0.4861111094554265,
+      "rewards/verify_correctness/std": 0.39317835370699566,
+      "step": 125
+    },
+    {
+      "clip_ratio": 0.007465552911162377,
+      "completion_length": 111.97916793823242,
+      "epoch": 0.26,
+      "grad_norm": 89.72521188925481,
+      "kl": 1.384375,
+      "learning_rate": 1e-06,
+      "loss": 0.0015,
+      "reward": 0.476250022649765,
+      "reward_std": 0.17444811761379242,
+      "rewards/judge_tool_use": 0.5197916775941849,
+      "rewards/judge_tool_use/std": 0.3664846569299698,
+      "rewards/verify_correctness": 0.3020833358168602,
+      "rewards/verify_correctness/std": 0.43010377883911133,
+      "step": 130
+    },
+    {
+      "clip_ratio": 0.006399596482515335,
+      "completion_length": 96.55555979410808,
+      "epoch": 0.27,
+      "grad_norm": 1.7399274702030754,
+      "kl": 0.966015625,
+      "learning_rate": 1e-06,
+      "loss": 0.0147,
+      "reward": 0.4886111219724019,
+      "reward_std": 0.23944897452990213,
+      "rewards/judge_tool_use": 0.5430555542310079,
+      "rewards/judge_tool_use/std": 0.3470989763736725,
+      "rewards/verify_correctness": 0.2708333420256774,
+      "rewards/verify_correctness/std": 0.37673963109652203,
+      "step": 135
+    },
+    {
+      "clip_ratio": 0.007890220731496811,
+      "completion_length": 113.35416793823242,
+      "epoch": 0.28,
+      "grad_norm": 2.231161419407326,
+      "kl": 3.628125,
+      "learning_rate": 1e-06,
+      "loss": 0.0151,
+      "reward": 0.38749998807907104,
+      "reward_std": 0.14481262117624283,
+      "rewards/judge_tool_use": 0.484375,
+      "rewards/judge_tool_use/std": 0.35279956459999084,
+      "rewards/verify_correctness": 0.0,
+      "rewards/verify_correctness/std": 0.0,
+      "step": 140
+    },
+    {
+      "clip_ratio": 0.004287280049175024,
+      "completion_length": 128.59722391764322,
+      "epoch": 0.29,
+      "grad_norm": 5.750983107539032,
+      "kl": 0.58671875,
+      "learning_rate": 1e-06,
+      "loss": 0.0151,
+      "reward": 0.42972222963968915,
+      "reward_std": 0.15125738829374313,
+      "rewards/judge_tool_use": 0.47291667262713116,
+      "rewards/judge_tool_use/std": 0.35496102770169574,
+      "rewards/verify_correctness": 0.2569444477558136,
+      "rewards/verify_correctness/std": 0.3278166850407918,
+      "step": 145
+    },
+    {
+      "clip_ratio": 0.007053372077643871,
+      "completion_length": 98.94791793823242,
+      "epoch": 0.3,
+      "grad_norm": 1.5954937607260369,
+      "kl": 0.237109375,
+      "learning_rate": 1e-06,
+      "loss": -0.0047,
+      "reward": 0.5891667008399963,
+      "reward_std": 0.14705681800842285,
+      "rewards/judge_tool_use": 0.6322916746139526,
+      "rewards/judge_tool_use/std": 0.37483371794223785,
+      "rewards/verify_correctness": 0.4166666567325592,
+      "rewards/verify_correctness/std": 0.49408674240112305,
+      "step": 150
+    },
+    {
+      "clip_ratio": 0.004191365744918585,
+      "completion_length": 88.18055979410808,
+      "epoch": 0.31,
+      "grad_norm": 2.1141011322952803,
+      "kl": 0.2513671875,
+      "learning_rate": 1e-06,
+      "loss": -0.002,
+      "reward": 0.7086111108462015,
+      "reward_std": 0.14623582363128662,
+      "rewards/judge_tool_use": 0.7486110925674438,
+      "rewards/judge_tool_use/std": 0.3543677230676015,
+      "rewards/verify_correctness": 0.5486110945542654,
+      "rewards/verify_correctness/std": 0.4976752698421478,
+      "step": 155
+    },
+    {
+      "clip_ratio": 0.007304594852030277,
+      "completion_length": 97.51041793823242,
+      "epoch": 0.32,
+      "grad_norm": 1.1578413021097405,
+      "kl": 0.332421875,
+      "learning_rate": 1e-06,
+      "loss": -0.0083,
+      "reward": 0.534166693687439,
+      "reward_std": 0.11162800714373589,
+      "rewards/judge_tool_use": 0.5843749940395355,
+      "rewards/judge_tool_use/std": 0.3592209666967392,
+      "rewards/verify_correctness": 0.3333333432674408,
+      "rewards/verify_correctness/std": 0.47639307379722595,
+      "step": 160
+    },
+    {
+      "clip_ratio": 0.0060629287734627725,
+      "completion_length": 110.75000254313152,
+      "epoch": 0.33,
+      "grad_norm": 0.8714902465835312,
+      "kl": 0.35859375,
+      "learning_rate": 1e-06,
+      "loss": 0.0151,
+      "reward": 0.5477778116861979,
+      "reward_std": 0.13536312182744345,
+      "rewards/judge_tool_use": 0.6013888915379842,
+      "rewards/judge_tool_use/std": 0.38596779108047485,
+      "rewards/verify_correctness": 0.3333333383003871,
+      "rewards/verify_correctness/std": 0.45276867349942523,
+      "step": 165
+    },
+    {
+      "clip_ratio": 0.005478941276669503,
+      "completion_length": 147.46875762939453,
+      "epoch": 0.34,
+      "grad_norm": 2.905066362863796,
+      "kl": 0.1998046875,
+      "learning_rate": 1e-06,
+      "loss": 0.0213,
+      "reward": 0.42500002682209015,
+      "reward_std": 0.17469991743564606,
+      "rewards/judge_tool_use": 0.4947916567325592,
+      "rewards/judge_tool_use/std": 0.346432626247406,
+      "rewards/verify_correctness": 0.1458333358168602,
+      "rewards/verify_correctness/std": 0.35542023181915283,
+      "step": 170
+    },
+    {
+      "clip_ratio": 0.003099467512220144,
+      "completion_length": 99.73611704508464,
+      "epoch": 0.35,
+      "grad_norm": 2.191551166799849,
+      "kl": 0.3734375,
+      "learning_rate": 1e-06,
+      "loss": 0.0089,
+      "reward": 0.7052778005599976,
+      "reward_std": 0.11197900461653869,
+      "rewards/judge_tool_use": 0.7652777632077535,
+      "rewards/judge_tool_use/std": 0.26592855403820675,
+      "rewards/verify_correctness": 0.4652777661879857,
+      "rewards/verify_correctness/std": 0.4105065166950226,
+      "step": 175
+    },
+    {
+      "clip_ratio": 0.005960878264158964,
+      "completion_length": 89.08333587646484,
+      "epoch": 0.36,
+      "grad_norm": 1.627066760583541,
+      "kl": 0.53984375,
+      "learning_rate": 1e-06,
+      "loss": 0.0001,
+      "reward": 0.6620833575725555,
+      "reward_std": 0.15964354574680328,
+      "rewards/judge_tool_use": 0.7312500178813934,
+      "rewards/judge_tool_use/std": 0.31979653239250183,
+      "rewards/verify_correctness": 0.3854166567325592,
+      "rewards/verify_correctness/std": 0.49077247083187103,
+      "step": 180
+    },
+    {
+      "clip_ratio": 0.004781427606940269,
+      "completion_length": 92.86805725097656,
+      "epoch": 0.37,
+      "grad_norm": 1.3401741313453097,
+      "kl": 0.434765625,
+      "learning_rate": 1e-06,
+      "loss": -0.0005,
+      "reward": 0.6336111426353455,
+      "reward_std": 0.10186164081096649,
+      "rewards/judge_tool_use": 0.6687500178813934,
+      "rewards/judge_tool_use/std": 0.3347410162289937,
+      "rewards/verify_correctness": 0.4930555423100789,
+      "rewards/verify_correctness/std": 0.4855302969614665,
+      "step": 185
+    },
+    {
+      "clip_ratio": 0.006658449023962021,
+      "completion_length": 108.47917175292969,
+      "epoch": 0.38,
+      "grad_norm": 2.5096982541940998,
+      "kl": 0.5453125,
+      "learning_rate": 1e-06,
+      "loss": -0.0017,
+      "reward": 0.5404166877269745,
+      "reward_std": 0.13422710448503494,
+      "rewards/judge_tool_use": 0.5895833522081375,
+      "rewards/judge_tool_use/std": 0.34020277857780457,
+      "rewards/verify_correctness": 0.34375,
+      "rewards/verify_correctness/std": 0.44986698031425476,
+      "step": 190
+    },
+    {
+      "clip_ratio": 0.004200516827404499,
+      "completion_length": 95.00694783528645,
+      "epoch": 0.39,
+      "grad_norm": 1.0766041890185425,
+      "kl": 0.31015625,
+      "learning_rate": 1e-06,
+      "loss": -0.0152,
+      "reward": 0.6466667056083679,
+      "reward_std": 0.18731083969275156,
+      "rewards/judge_tool_use": 0.6972222328186035,
+      "rewards/judge_tool_use/std": 0.3268563648064931,
+      "rewards/verify_correctness": 0.4444444477558136,
+      "rewards/verify_correctness/std": 0.4707372883955638,
+      "step": 195
+    },
+    {
+      "clip_ratio": 0.005295699182897806,
+      "completion_length": 122.48958587646484,
+      "epoch": 0.4,
+      "grad_norm": 2.4546002318741573,
+      "kl": 0.50234375,
+      "learning_rate": 1e-06,
+      "loss": 0.0118,
+      "reward": 0.6545833051204681,
+      "reward_std": 0.13637181371450424,
+      "rewards/judge_tool_use": 0.7531250417232513,
+      "rewards/judge_tool_use/std": 0.27382777631282806,
+      "rewards/verify_correctness": 0.2604166641831398,
+      "rewards/verify_correctness/std": 0.42997145652770996,
+      "step": 200
+    },
+    {
+      "clip_ratio": 0.0038763705641031267,
+      "completion_length": 92.8402811686198,
+      "epoch": 0.41,
+      "grad_norm": 1.07948303423069,
+      "kl": 0.391796875,
+      "learning_rate": 1e-06,
+      "loss": -0.0001,
+      "reward": 0.7358333865801493,
+      "reward_std": 0.14563809831937155,
+      "rewards/judge_tool_use": 0.7756944497426351,
+      "rewards/judge_tool_use/std": 0.3286278744538625,
+      "rewards/verify_correctness": 0.5763888955116272,
+      "rewards/verify_correctness/std": 0.4962470730145772,
+      "step": 205
+    },
+    {
+      "clip_ratio": 0.005373115837574005,
+      "completion_length": 149.50000762939453,
+      "epoch": 0.42,
+      "grad_norm": 2.7654236631039004,
+      "kl": 0.3029296875,
+      "learning_rate": 1e-06,
+      "loss": 0.006,
+      "reward": 0.6633333265781403,
+      "reward_std": 0.14661981165409088,
+      "rewards/judge_tool_use": 0.7197916805744171,
+      "rewards/judge_tool_use/std": 0.3548731952905655,
+      "rewards/verify_correctness": 0.4374999850988388,
+      "rewards/verify_correctness/std": 0.49408672749996185,
+      "step": 210
+    },
+    {
+      "clip_ratio": 0.0028386844322085382,
+      "completion_length": 108.20833333333333,
+      "epoch": 0.43,
+      "grad_norm": 2.009171248811446,
+      "kl": 0.340234375,
+      "learning_rate": 1e-06,
+      "loss": -0.0052,
+      "reward": 0.7725000381469727,
+      "reward_std": 0.09887611617644627,
+      "rewards/judge_tool_use": 0.8493055502573649,
+      "rewards/judge_tool_use/std": 0.27945346136887866,
+      "rewards/verify_correctness": 0.4652777910232544,
+      "rewards/verify_correctness/std": 0.47633983691533405,
+      "step": 215
+    },
+    {
+      "clip_ratio": 0.0061874065548181536,
+      "completion_length": 117.18750381469727,
+      "epoch": 0.44,
+      "grad_norm": 6.685116857049659,
+      "kl": 0.394140625,
+      "learning_rate": 1e-06,
+      "loss": 0.0106,
+      "reward": 0.6800000071525574,
+      "reward_std": 0.19070632010698318,
+      "rewards/judge_tool_use": 0.7510416507720947,
+      "rewards/judge_tool_use/std": 0.3492521345615387,
+      "rewards/verify_correctness": 0.3958333283662796,
+      "rewards/verify_correctness/std": 0.4937349408864975,
+      "step": 220
+    },
+    {
+      "clip_ratio": 0.0023891992866992952,
+      "completion_length": 105.63195037841797,
+      "epoch": 0.45,
+      "grad_norm": 0.639993744814138,
+      "kl": 0.55,
+      "learning_rate": 1e-06,
+      "loss": 0.0022,
+      "reward": 0.7469444473584493,
+      "reward_std": 0.11130555470784505,
+      "rewards/judge_tool_use": 0.7756944298744202,
+      "rewards/judge_tool_use/std": 0.3260296682516734,
+      "rewards/verify_correctness": 0.6319444378217062,
+      "rewards/verify_correctness/std": 0.47723113497098285,
+      "step": 225
+    },
+    {
+      "clip_ratio": 0.004893714282661677,
+      "completion_length": 106.87500381469727,
+      "epoch": 0.46,
+      "grad_norm": 1.994854766362101,
+      "kl": 0.412890625,
+      "learning_rate": 1e-06,
+      "loss": 0.0088,
+      "reward": 0.7974999845027924,
+      "reward_std": 0.11849463731050491,
+      "rewards/judge_tool_use": 0.8666666746139526,
+      "rewards/judge_tool_use/std": 0.22190550714731216,
+      "rewards/verify_correctness": 0.5208333432674408,
+      "rewards/verify_correctness/std": 0.49776600301265717,
+      "step": 230
+    },
+    {
+      "clip_ratio": 0.0027995842508971693,
+      "completion_length": 98.25000254313152,
+      "epoch": 0.47,
+      "grad_norm": 0.8370010567585779,
+      "kl": 0.578515625,
+      "learning_rate": 1e-06,
+      "loss": 0.0069,
+      "reward": 0.7361111442248026,
+      "reward_std": 0.08546942720810573,
+      "rewards/judge_tool_use": 0.7708333333333334,
+      "rewards/judge_tool_use/std": 0.3326546351114909,
+      "rewards/verify_correctness": 0.597222218910853,
+      "rewards/verify_correctness/std": 0.48818884293238324,
+      "step": 235
+    },
+    {
+      "clip_ratio": 0.0028807530645281075,
+      "completion_length": 127.83333587646484,
+      "epoch": 0.48,
+      "grad_norm": 1.0910338182282877,
+      "kl": 0.3333984375,
+      "learning_rate": 1e-06,
+      "loss": 0.0078,
+      "reward": 0.8179166615009308,
+      "reward_std": 0.10780513286590576,
+      "rewards/judge_tool_use": 0.8739583194255829,
+      "rewards/judge_tool_use/std": 0.2250625640153885,
+      "rewards/verify_correctness": 0.5937499850988388,
+      "rewards/verify_correctness/std": 0.48209600150585175,
+      "step": 240
+    },
+    {
+      "clip_ratio": 0.002781028486788273,
+      "completion_length": 143.06944783528647,
+      "epoch": 0.49,
+      "grad_norm": 0.8178970498503804,
+      "kl": 0.7380859375,
+      "learning_rate": 1e-06,
+      "loss": 0.0001,
+      "reward": 0.6788889169692993,
+      "reward_std": 0.10102842872341473,
+      "rewards/judge_tool_use": 0.7027777830759684,
+      "rewards/judge_tool_use/std": 0.3571552236874898,
+      "rewards/verify_correctness": 0.5833333432674408,
+      "rewards/verify_correctness/std": 0.4730796217918396,
+      "step": 245
+    },
+    {
+      "clip_ratio": 0.004690852202475071,
+      "completion_length": 127.69792175292969,
+      "epoch": 0.5,
+      "grad_norm": 0.7032026926016068,
+      "kl": 0.434375,
+      "learning_rate": 1e-06,
+      "loss": 0.0056,
+      "reward": 0.7641666829586029,
+      "reward_std": 0.1450955532491207,
+      "rewards/judge_tool_use": 0.8354166746139526,
+      "rewards/judge_tool_use/std": 0.2628621906042099,
+      "rewards/verify_correctness": 0.4791666567325592,
+      "rewards/verify_correctness/std": 0.5048523545265198,
+      "step": 250
+    },
+    {
+      "clip_ratio": 0.003919628076255321,
+      "completion_length": 110.83333587646484,
+      "epoch": 0.51,
+      "grad_norm": 1.7992035669287993,
+      "kl": 0.4609375,
+      "learning_rate": 1e-06,
+      "loss": 0.0157,
+      "reward": 0.7619444529215494,
+      "reward_std": 0.14088333398103714,
+      "rewards/judge_tool_use": 0.8222222725550333,
+      "rewards/judge_tool_use/std": 0.30879923701286316,
+      "rewards/verify_correctness": 0.5208333333333334,
+      "rewards/verify_correctness/std": 0.48915791511535645,
+      "step": 255
+    },
+    {
+      "clip_ratio": 0.005749301984906197,
+      "completion_length": 112.65625,
+      "epoch": 0.52,
+      "grad_norm": 3.1299065923806113,
+      "kl": 1.12109375,
+      "learning_rate": 1e-06,
+      "loss": 0.001,
+      "reward": 0.7679167091846466,
+      "reward_std": 0.10637958347797394,
+      "rewards/judge_tool_use": 0.8374999761581421,
+      "rewards/judge_tool_use/std": 0.29924115538597107,
+      "rewards/verify_correctness": 0.4895833283662796,
+      "rewards/verify_correctness/std": 0.505071759223938,
+      "step": 260
+    },
+    {
+      "clip_ratio": 0.003776553086936474,
+      "completion_length": 114.35417175292969,
+      "epoch": 0.53,
+      "grad_norm": 0.6562019375611381,
+      "kl": 0.563671875,
+      "learning_rate": 1e-06,
+      "loss": 0.0019,
+      "reward": 0.7005555629730225,
+      "reward_std": 0.15318704272309938,
+      "rewards/judge_tool_use": 0.7854166825612386,
+      "rewards/judge_tool_use/std": 0.3270541826883952,
+      "rewards/verify_correctness": 0.3611111218730609,
+      "rewards/verify_correctness/std": 0.4106987814108531,
+      "step": 265
+    },
+    {
+      "clip_ratio": 0.0034743762575089932,
+      "completion_length": 147.875,
+      "epoch": 0.54,
+      "grad_norm": 1.3887743953927447,
+      "kl": 0.3212890625,
+      "learning_rate": 1e-06,
+      "loss": 0.0034,
+      "reward": 0.8141666948795319,
+      "reward_std": 0.12217209488153458,
+      "rewards/judge_tool_use": 0.8979166746139526,
+      "rewards/judge_tool_use/std": 0.16565020941197872,
+      "rewards/verify_correctness": 0.4791666716337204,
+      "rewards/verify_correctness/std": 0.49776598811149597,
+      "step": 270
+    },
+    {
+      "clip_ratio": 0.003613210655748844,
+      "completion_length": 105.86805725097656,
+      "epoch": 0.55,
+      "grad_norm": 0.9278154163608585,
+      "kl": 1.162890625,
+      "learning_rate": 1e-06,
+      "loss": 0.0073,
+      "reward": 0.7766666611035665,
+      "reward_std": 0.16950395703315735,
+      "rewards/judge_tool_use": 0.8145833412806193,
+      "rewards/judge_tool_use/std": 0.3171912630399068,
+      "rewards/verify_correctness": 0.625,
+      "rewards/verify_correctness/std": 0.48924607038497925,
+      "step": 275
+    },
+    {
+      "clip_ratio": 0.005631557852029801,
+      "completion_length": 88.15625,
+      "epoch": 0.56,
+      "grad_norm": 0.49513702069352056,
+      "kl": 0.349609375,
+      "learning_rate": 1e-06,
+      "loss": 0.0046,
+      "reward": 0.9312500357627869,
+      "reward_std": 0.0591347049921751,
+      "rewards/judge_tool_use": 0.9583333134651184,
+      "rewards/judge_tool_use/std": 0.11368590220808983,
+      "rewards/verify_correctness": 0.8229166567325592,
+      "rewards/verify_correctness/std": 0.38553228974342346,
+      "step": 280
+    },
+    {
+      "clip_ratio": 0.0010083256289362907,
+      "completion_length": 111.0277811686198,
+      "epoch": 0.57,
+      "grad_norm": 0.9774895117172071,
+      "kl": 0.299609375,
+      "learning_rate": 1e-06,
+      "loss": 0.0043,
+      "reward": 0.8247222304344177,
+      "reward_std": 0.12139152487119038,
+      "rewards/judge_tool_use": 0.8694444497426351,
+      "rewards/judge_tool_use/std": 0.2646999955177307,
+      "rewards/verify_correctness": 0.6458333333333334,
+      "rewards/verify_correctness/std": 0.4829867482185364,
+      "step": 285
+    },
+    {
+      "clip_ratio": 0.004534756857901811,
+      "completion_length": 138.50000381469727,
+      "epoch": 0.58,
+      "grad_norm": 1.430617019290729,
+      "kl": 0.3974609375,
+      "learning_rate": 1e-06,
+      "loss": 0.0021,
+      "reward": 0.64000004529953,
+      "reward_std": 0.129538856446743,
+      "rewards/judge_tool_use": 0.7166666686534882,
+      "rewards/judge_tool_use/std": 0.3585747927427292,
+      "rewards/verify_correctness": 0.3333333358168602,
+      "rewards/verify_correctness/std": 0.4409564733505249,
+      "step": 290
+    },
+    {
+      "clip_ratio": 0.003662421926856041,
+      "completion_length": 118.62500254313152,
+      "epoch": 0.59,
+      "grad_norm": 0.779481967016939,
+      "kl": 0.531640625,
+      "learning_rate": 1e-06,
+      "loss": 0.0063,
+      "reward": 0.8033333619435629,
+      "reward_std": 0.13420975704987845,
+      "rewards/judge_tool_use": 0.8756944338480631,
+      "rewards/judge_tool_use/std": 0.23473882675170898,
+      "rewards/verify_correctness": 0.5138888855775198,
+      "rewards/verify_correctness/std": 0.40222856402397156,
+      "step": 295
+    },
+    {
+      "clip_ratio": 0.0028870035894215107,
+      "completion_length": 113.86458587646484,
+      "epoch": 0.6,
+      "grad_norm": 0.8439122527617477,
+      "kl": 0.2859375,
+      "learning_rate": 1e-06,
+      "loss": -0.0006,
+      "reward": 0.8366667032241821,
+      "reward_std": 0.07533844374120235,
+      "rewards/judge_tool_use": 0.8999999761581421,
+      "rewards/judge_tool_use/std": 0.2058699131011963,
+      "rewards/verify_correctness": 0.5833333432674408,
+      "rewards/verify_correctness/std": 0.4908421188592911,
+      "step": 300
+    },
+    {
+      "clip_ratio": 0.0034239688422530888,
+      "completion_length": 106.85416920979817,
+      "epoch": 0.61,
+      "grad_norm": 2.015558621646506,
+      "kl": 0.965625,
+      "learning_rate": 1e-06,
+      "loss": 0.0182,
+      "reward": 0.8475000063578287,
+      "reward_std": 0.1549657036860784,
+      "rewards/judge_tool_use": 0.8909722169240316,
+      "rewards/judge_tool_use/std": 0.24881805976231894,
+      "rewards/verify_correctness": 0.6736111044883728,
+      "rewards/verify_correctness/std": 0.43851491808891296,
+      "step": 305
+    },
+    {
+      "clip_ratio": 0.0033080252818763257,
+      "completion_length": 135.61459350585938,
+      "epoch": 0.62,
+      "grad_norm": 2.3005763258000322,
+      "kl": 1.21640625,
+      "learning_rate": 1e-06,
+      "loss": 0.0275,
+      "reward": 0.8000000417232513,
+      "reward_std": 0.14312875270843506,
+      "rewards/judge_tool_use": 0.8281249403953552,
+      "rewards/judge_tool_use/std": 0.29270513355731964,
+      "rewards/verify_correctness": 0.6875,
+      "rewards/verify_correctness/std": 0.4634205251932144,
+      "step": 310
+    },
+    {
+      "clip_ratio": 0.0017321385443210603,
+      "completion_length": 103.0763931274414,
+      "epoch": 0.63,
+      "grad_norm": 80.49583376546836,
+      "kl": 4.99609375,
+      "learning_rate": 1e-06,
+      "loss": 0.0215,
+      "reward": 0.9177778164545695,
+      "reward_std": 0.07255559042096138,
+      "rewards/judge_tool_use": 0.9527777433395386,
+      "rewards/judge_tool_use/std": 0.12888848781585693,
+      "rewards/verify_correctness": 0.7777777711550394,
+      "rewards/verify_correctness/std": 0.40553895632425946,
+      "step": 315
+    },
+    {
+      "clip_ratio": 0.0032403219491243364,
+      "completion_length": 97.76042175292969,
+      "epoch": 0.64,
+      "grad_norm": 2.6596636763815953,
+      "kl": 1.287109375,
+      "learning_rate": 1e-06,
+      "loss": 0.011,
+      "reward": 0.8733333945274353,
+      "reward_std": 0.08747303485870361,
+      "rewards/judge_tool_use": 0.9145833253860474,
+      "rewards/judge_tool_use/std": 0.20979993045330048,
+      "rewards/verify_correctness": 0.7083333432674408,
+      "rewards/verify_correctness/std": 0.45699402689933777,
+      "step": 320
+    },
+    {
+      "clip_ratio": 0.0033186592161655428,
+      "completion_length": 117.89583841959636,
+      "epoch": 0.65,
+      "grad_norm": 1.1152329582646996,
+      "kl": 0.5712890625,
+      "learning_rate": 1e-06,
+      "loss": 0.02,
+      "reward": 0.7755555907885233,
+      "reward_std": 0.13340077300866446,
+      "rewards/judge_tool_use": 0.7993055383364359,
+      "rewards/judge_tool_use/std": 0.28823333978652954,
+      "rewards/verify_correctness": 0.6805555621782938,
+      "rewards/verify_correctness/std": 0.4147060612837474,
+      "step": 325
+    },
+    {
+      "clip_ratio": 0.0034143137745559216,
+      "completion_length": 83.44791793823242,
+      "epoch": 0.66,
+      "grad_norm": 0.9507226451435283,
+      "kl": 0.600390625,
+      "learning_rate": 1e-06,
+      "loss": 0.0022,
+      "reward": 0.9216666221618652,
+      "reward_std": 0.0693933442234993,
+      "rewards/judge_tool_use": 0.9593749642372131,
+      "rewards/judge_tool_use/std": 0.10937883332371712,
+      "rewards/verify_correctness": 0.7708333134651184,
+      "rewards/verify_correctness/std": 0.3960151970386505,
+      "step": 330
+    },
+    {
+      "clip_ratio": 0.00203145882114768,
+      "completion_length": 143.0763956705729,
+      "epoch": 0.67,
+      "grad_norm": 1.94906459086153,
+      "kl": 0.6546875,
+      "learning_rate": 1e-06,
+      "loss": 0.0089,
+      "reward": 0.820555567741394,
+      "reward_std": 0.10365821545322736,
+      "rewards/judge_tool_use": 0.8833333253860474,
+      "rewards/judge_tool_use/std": 0.21480779349803925,
+      "rewards/verify_correctness": 0.5694444477558136,
+      "rewards/verify_correctness/std": 0.49419113993644714,
+      "step": 335
+    },
+    {
+      "clip_ratio": 0.005462261289358139,
+      "completion_length": 96.75,
+      "epoch": 0.68,
+      "grad_norm": 3.6005290276382365,
+      "kl": 1.08046875,
+      "learning_rate": 1e-06,
+      "loss": 0.0146,
+      "reward": 0.84333336353302,
+      "reward_std": 0.10394849628210068,
+      "rewards/judge_tool_use": 0.8562499582767487,
+      "rewards/judge_tool_use/std": 0.28872600197792053,
+      "rewards/verify_correctness": 0.7916666567325592,
+      "rewards/verify_correctness/std": 0.3885633647441864,
+      "step": 340
+    },
+    {
+      "clip_ratio": 0.00261396998539567,
+      "completion_length": 139.52083841959634,
+      "epoch": 0.69,
+      "grad_norm": 2.595391200258739,
+      "kl": 0.625,
+      "learning_rate": 1e-06,
+      "loss": 0.0304,
+      "reward": 0.7783333659172058,
+      "reward_std": 0.16342870394388834,
+      "rewards/judge_tool_use": 0.8201388716697693,
+      "rewards/judge_tool_use/std": 0.27010903755823773,
+      "rewards/verify_correctness": 0.6111111144224802,
+      "rewards/verify_correctness/std": 0.46264639496803284,
+      "step": 345
+    },
+    {
+      "clip_ratio": 0.005096960160881281,
+      "completion_length": 109.18750381469727,
+      "epoch": 0.7,
+      "grad_norm": 2.509616050978152,
+      "kl": 1.11640625,
+      "learning_rate": 1e-06,
+      "loss": 0.0095,
+      "reward": 0.8020833134651184,
+      "reward_std": 0.1078796535730362,
+      "rewards/judge_tool_use": 0.84375,
+      "rewards/judge_tool_use/std": 0.23881300538778305,
+      "rewards/verify_correctness": 0.6354166716337204,
+      "rewards/verify_correctness/std": 0.4489881992340088,
+      "step": 350
+    },
+    {
+      "clip_ratio": 0.0030191333964467047,
+      "completion_length": 119.64583587646484,
+      "epoch": 0.71,
+      "grad_norm": 1.9232078822128884,
+      "kl": 0.651171875,
+      "learning_rate": 1e-06,
+      "loss": 0.0193,
+      "reward": 0.8113889296849569,
+      "reward_std": 0.09605963776508968,
+      "rewards/judge_tool_use": 0.8597222169240316,
+      "rewards/judge_tool_use/std": 0.23875926931699118,
+      "rewards/verify_correctness": 0.6180555621782938,
+      "rewards/verify_correctness/std": 0.41723007957140607,
+      "step": 355
+    },
+    {
+      "clip_ratio": 0.0040064876899123195,
+      "completion_length": 107.82292175292969,
+      "epoch": 0.72,
+      "grad_norm": 0.8384093480805856,
+      "kl": 0.59375,
+      "learning_rate": 1e-06,
+      "loss": 0.0015,
+      "reward": 0.8174999952316284,
+      "reward_std": 0.07755549252033234,
+      "rewards/judge_tool_use": 0.8760416805744171,
+      "rewards/judge_tool_use/std": 0.21128008142113686,
+      "rewards/verify_correctness": 0.5833333432674408,
+      "rewards/verify_correctness/std": 0.4908421188592911,
+      "step": 360
+    },
+    {
+      "clip_ratio": 0.003883673809468746,
+      "completion_length": 118.04167175292969,
+      "epoch": 0.73,
+      "grad_norm": 3.6270547715534547,
+      "kl": 0.630859375,
+      "learning_rate": 1e-06,
+      "loss": 0.0123,
+      "reward": 0.8119444648424784,
+      "reward_std": 0.10799338171879451,
+      "rewards/judge_tool_use": 0.8708332975705465,
+      "rewards/judge_tool_use/std": 0.24002541601657867,
+      "rewards/verify_correctness": 0.5763889054457346,
+      "rewards/verify_correctness/std": 0.49137688676516217,
+      "step": 365
+    },
+    {
+      "clip_ratio": 0.002587809320539236,
+      "completion_length": 120.89583587646484,
+      "epoch": 0.74,
+      "grad_norm": 2.313493437394313,
+      "kl": 0.56640625,
+      "learning_rate": 1e-06,
+      "loss": 0.0077,
+      "reward": 0.8029167056083679,
+      "reward_std": 0.12100144848227501,
+      "rewards/judge_tool_use": 0.8760416805744171,
+      "rewards/judge_tool_use/std": 0.26075808703899384,
+      "rewards/verify_correctness": 0.5104166716337204,
+      "rewards/verify_correctness/std": 0.47240525484085083,
+      "step": 370
+    },
+    {
+      "clip_ratio": 0.0027662038803100584,
+      "completion_length": 130.86805979410806,
+      "epoch": 0.75,
+      "grad_norm": 1.4291418174291879,
+      "kl": 0.58203125,
+      "learning_rate": 1e-06,
+      "loss": 0.0109,
+      "reward": 0.859166661898295,
+      "reward_std": 0.13911124070485434,
+      "rewards/judge_tool_use": 0.8986111283302307,
+      "rewards/judge_tool_use/std": 0.21422065049409866,
+      "rewards/verify_correctness": 0.7013888756434122,
+      "rewards/verify_correctness/std": 0.4320597251256307,
+      "step": 375
+    },
+    {
+      "clip_ratio": 0.004125702893361449,
+      "completion_length": 102.20833587646484,
+      "epoch": 0.76,
+      "grad_norm": 0.8091042916650757,
+      "kl": 0.345703125,
+      "learning_rate": 1e-06,
+      "loss": 0.001,
+      "reward": 0.8775000274181366,
+      "reward_std": 0.07746638730168343,
+      "rewards/judge_tool_use": 0.925000011920929,
+      "rewards/judge_tool_use/std": 0.19297337904572487,
+      "rewards/verify_correctness": 0.6875,
+      "rewards/verify_correctness/std": 0.4594741761684418,
+      "step": 380
+    },
+    {
+      "clip_ratio": 0.0013461930211633444,
+      "completion_length": 116.84722391764323,
+      "epoch": 0.77,
+      "grad_norm": 0.8881903330505401,
+      "kl": 3.155859375,
+      "learning_rate": 1e-06,
+      "loss": 0.0164,
+      "reward": 0.8816666801770529,
+      "reward_std": 0.07364911710222562,
+      "rewards/judge_tool_use": 0.9284722208976746,
+      "rewards/judge_tool_use/std": 0.1873454600572586,
+      "rewards/verify_correctness": 0.694444457689921,
+      "rewards/verify_correctness/std": 0.4634603758653005,
+      "step": 385
+    },
+    {
+      "clip_ratio": 0.0035483752377331256,
+      "completion_length": 113.29166793823242,
+      "epoch": 0.78,
+      "grad_norm": 1.9115550749698071,
+      "kl": 0.36015625,
+      "learning_rate": 1e-06,
+      "loss": 0.006,
+      "reward": 0.9200000166893005,
+      "reward_std": 0.07504565827548504,
+      "rewards/judge_tool_use": 0.9677083492279053,
+      "rewards/judge_tool_use/std": 0.0825247336179018,
+      "rewards/verify_correctness": 0.7291666865348816,
+      "rewards/verify_correctness/std": 0.44340358674526215,
+      "step": 390
+    },
+    {
+      "clip_ratio": 0.0033025143668055533,
+      "completion_length": 118.38194529215495,
+      "epoch": 0.79,
+      "grad_norm": 0.6918361378607686,
+      "kl": 0.459375,
+      "learning_rate": 1e-06,
+      "loss": 0.0016,
+      "reward": 0.8894444505373637,
+      "reward_std": 0.09632904827594757,
+      "rewards/judge_tool_use": 0.9416666428248087,
+      "rewards/judge_tool_use/std": 0.13143713772296906,
+      "rewards/verify_correctness": 0.6805555621782938,
+      "rewards/verify_correctness/std": 0.46167699495951336,
+      "step": 395
+    },
+    {
+      "clip_ratio": 0.003618121985346079,
+      "completion_length": 118.16666793823242,
+      "epoch": 0.8,
+      "grad_norm": 1.7518446245486312,
+      "kl": 0.746484375,
+      "learning_rate": 1e-06,
+      "loss": 0.0075,
+      "reward": 0.8358333706855774,
+      "reward_std": 0.1284572072327137,
+      "rewards/judge_tool_use": 0.8625000417232513,
+      "rewards/judge_tool_use/std": 0.27199896425008774,
+      "rewards/verify_correctness": 0.7291666567325592,
+      "rewards/verify_correctness/std": 0.4388819485902786,
+      "step": 400
+    },
+    {
+      "clip_ratio": 0.003691716957837343,
+      "completion_length": 110.00694783528645,
+      "epoch": 0.81,
+      "grad_norm": 0.8896421929804095,
+      "kl": 0.3625,
+      "learning_rate": 1e-06,
+      "loss": 0.0004,
+      "reward": 0.8361111481984457,
+      "reward_std": 0.08633811896046002,
+      "rewards/judge_tool_use": 0.8958333333333334,
+      "rewards/judge_tool_use/std": 0.1966320425271988,
+      "rewards/verify_correctness": 0.5972222089767456,
+      "rewards/verify_correctness/std": 0.48761820793151855,
+      "step": 405
+    },
+    {
+      "clip_ratio": 0.003539162687957287,
+      "completion_length": 114.38542175292969,
+      "epoch": 0.82,
+      "grad_norm": 0.8364119133705628,
+      "kl": 0.416796875,
+      "learning_rate": 1e-06,
+      "loss": 0.0046,
+      "reward": 0.8329166769981384,
+      "reward_std": 0.07315276563167572,
+      "rewards/judge_tool_use": 0.8770833611488342,
+      "rewards/judge_tool_use/std": 0.2352844625711441,
+      "rewards/verify_correctness": 0.65625,
+      "rewards/verify_correctness/std": 0.44986698031425476,
+      "step": 410
+    },
+    {
+      "clip_ratio": 0.001082425657659769,
+      "completion_length": 105.50694783528645,
+      "epoch": 0.83,
+      "grad_norm": 1.1930359747277361,
+      "kl": 0.505078125,
+      "learning_rate": 1e-06,
+      "loss": 0.0135,
+      "reward": 0.8663889169692993,
+      "reward_std": 0.08535568416118622,
+      "rewards/judge_tool_use": 0.9388888875643412,
+      "rewards/judge_tool_use/std": 0.1443981478611628,
+      "rewards/verify_correctness": 0.5763888855775198,
+      "rewards/verify_correctness/std": 0.4668743809064229,
+      "step": 415
+    },
+    {
+      "clip_ratio": 0.0036332331597805022,
+      "completion_length": 91.27083587646484,
+      "epoch": 0.84,
+      "grad_norm": 1.0418583888118527,
+      "kl": 0.48046875,
+      "learning_rate": 1e-06,
+      "loss": 0.0021,
+      "reward": 0.8641667366027832,
+      "reward_std": 0.08411096408963203,
+      "rewards/judge_tool_use": 0.9135416746139526,
+      "rewards/judge_tool_use/std": 0.1935550421476364,
+      "rewards/verify_correctness": 0.6666666567325592,
+      "rewards/verify_correctness/std": 0.44964754581451416,
+      "step": 420
+    },
+    {
+      "clip_ratio": 0.0031434688717126846,
+      "completion_length": 136.47222900390625,
+      "epoch": 0.85,
+      "grad_norm": 9.602708864091571,
+      "kl": 0.8896484375,
+      "learning_rate": 1e-06,
+      "loss": 0.0095,
+      "reward": 0.8213888804117838,
+      "reward_std": 0.10578912993272145,
+      "rewards/judge_tool_use": 0.8791666825612386,
+      "rewards/judge_tool_use/std": 0.21359455088774362,
+      "rewards/verify_correctness": 0.5902777711550394,
+      "rewards/verify_correctness/std": 0.48865292469660443,
+      "step": 425
+    },
+    {
+      "clip_ratio": 0.006819223985075951,
+      "completion_length": 121.85417175292969,
+      "epoch": 0.86,
+      "grad_norm": 0.9453508121139471,
+      "kl": 0.821875,
+      "learning_rate": 1e-06,
+      "loss": 0.0144,
+      "reward": 0.7583333849906921,
+      "reward_std": 0.15919847041368484,
+      "rewards/judge_tool_use": 0.8177083134651184,
+      "rewards/judge_tool_use/std": 0.27655909955501556,
+      "rewards/verify_correctness": 0.5208333283662796,
+      "rewards/verify_correctness/std": 0.5030901730060577,
+      "step": 430
+    },
+    {
+      "clip_ratio": 0.005021798610687256,
+      "completion_length": 117.34722391764323,
+      "epoch": 0.87,
+      "grad_norm": 0.9574949482570072,
+      "kl": 2.016796875,
+      "learning_rate": 1e-06,
+      "loss": 0.0024,
+      "reward": 0.8338889280954996,
+      "reward_std": 0.11587712913751602,
+      "rewards/judge_tool_use": 0.868749996026357,
+      "rewards/judge_tool_use/std": 0.24017359067996344,
+      "rewards/verify_correctness": 0.6944444378217062,
+      "rewards/verify_correctness/std": 0.46080262462298077,
+      "step": 435
+    },
+    {
+      "clip_ratio": 0.003334458451718092,
+      "completion_length": 102.16667175292969,
+      "epoch": 0.88,
+      "grad_norm": 1.284555858192564,
+      "kl": 0.491015625,
+      "learning_rate": 1e-06,
+      "loss": 0.0001,
+      "reward": 0.8824999928474426,
+      "reward_std": 0.07597887888550758,
+      "rewards/judge_tool_use": 0.9364582896232605,
+      "rewards/judge_tool_use/std": 0.18419649451971054,
+      "rewards/verify_correctness": 0.6666666865348816,
+      "rewards/verify_correctness/std": 0.47639307379722595,
+      "step": 440
+    },
+    {
+      "clip_ratio": 0.0027906086295843126,
+      "completion_length": 98.38194783528645,
+      "epoch": 0.89,
+      "grad_norm": 0.4954906916589266,
+      "kl": 0.650390625,
+      "learning_rate": 1e-06,
+      "loss": 0.0002,
+      "reward": 0.8658333619435629,
+      "reward_std": 0.07699522500236829,
+      "rewards/judge_tool_use": 0.9381944338480631,
+      "rewards/judge_tool_use/std": 0.132844394693772,
+      "rewards/verify_correctness": 0.5763888955116272,
+      "rewards/verify_correctness/std": 0.47234441836675006,
+      "step": 445
+    },
+    {
+      "clip_ratio": 0.003258515940979123,
+      "completion_length": 87.12500381469727,
+      "epoch": 0.9,
+      "grad_norm": 0.8195724812478982,
+      "kl": 0.283984375,
+      "learning_rate": 1e-06,
+      "loss": -0.0011,
+      "reward": 0.8491666615009308,
+      "reward_std": 0.10520510375499725,
+      "rewards/judge_tool_use": 0.9000000357627869,
+      "rewards/judge_tool_use/std": 0.19222086668014526,
+      "rewards/verify_correctness": 0.6458333283662796,
+      "rewards/verify_correctness/std": 0.44007769227027893,
+      "step": 450
+    },
+    {
+      "clip_ratio": 0.003890740033239126,
+      "completion_length": 109.83333587646484,
+      "epoch": 0.91,
+      "grad_norm": 0.7607892651828614,
+      "kl": 0.4947265625,
+      "learning_rate": 1e-06,
+      "loss": -0.0018,
+      "reward": 0.8672222892443339,
+      "reward_std": 0.11101956913868587,
+      "rewards/judge_tool_use": 0.8999999761581421,
+      "rewards/judge_tool_use/std": 0.19080724567174911,
+      "rewards/verify_correctness": 0.7361111243565878,
+      "rewards/verify_correctness/std": 0.43669142325719196,
+      "step": 455
+    },
+    {
+      "clip_ratio": 0.003977376967668533,
+      "completion_length": 104.58333969116211,
+      "epoch": 0.92,
+      "grad_norm": 0.31311068474329323,
+      "kl": 0.4921875,
+      "learning_rate": 1e-06,
+      "loss": 0.0072,
+      "reward": 0.8212500214576721,
+      "reward_std": 0.08914041426032782,
+      "rewards/judge_tool_use": 0.8364583849906921,
+      "rewards/judge_tool_use/std": 0.2007633000612259,
+      "rewards/verify_correctness": 0.7604166567325592,
+      "rewards/verify_correctness/std": 0.3714234232902527,
+      "step": 460
+    },
+    {
+      "clip_ratio": 0.0033224607817828656,
+      "completion_length": 109.66666920979817,
+      "epoch": 0.93,
+      "grad_norm": 0.9597186134421405,
+      "kl": 0.775,
+      "learning_rate": 1e-06,
+      "loss": 0.003,
+      "reward": 0.7788889010747274,
+      "reward_std": 0.07402739052971204,
+      "rewards/judge_tool_use": 0.8104166587193807,
+      "rewards/judge_tool_use/std": 0.24389235799511275,
+      "rewards/verify_correctness": 0.652777781089147,
+      "rewards/verify_correctness/std": 0.3849090288082759,
+      "step": 465
+    },
+    {
+      "clip_ratio": 0.004331017844378948,
+      "completion_length": 128.03125762939453,
+      "epoch": 0.94,
+      "grad_norm": 0.6604515564011074,
+      "kl": 0.440234375,
+      "learning_rate": 1e-06,
+      "loss": 0.002,
+      "reward": 0.8320833444595337,
+      "reward_std": 0.11683276295661926,
+      "rewards/judge_tool_use": 0.8968749940395355,
+      "rewards/judge_tool_use/std": 0.20512499660253525,
+      "rewards/verify_correctness": 0.5729166567325592,
+      "rewards/verify_correctness/std": 0.49430613219738007,
+      "step": 470
+    },
+    {
+      "clip_ratio": 0.003034461010247469,
+      "completion_length": 129.48611704508463,
+      "epoch": 0.95,
+      "grad_norm": 1.7164077020687818,
+      "kl": 0.575390625,
+      "learning_rate": 1e-06,
+      "loss": 0.0071,
+      "reward": 0.7255556186040243,
+      "reward_std": 0.09651681656638782,
+      "rewards/judge_tool_use": 0.7993055780728658,
+      "rewards/judge_tool_use/std": 0.2533370802799861,
+      "rewards/verify_correctness": 0.4305555572112401,
+      "rewards/verify_correctness/std": 0.41316067179044086,
+      "step": 475
+    },
+    {
+      "clip_ratio": 0.004247188335284591,
+      "completion_length": 90.69791793823242,
+      "epoch": 0.96,
+      "grad_norm": 0.26610601551372964,
+      "kl": 0.3984375,
+      "learning_rate": 1e-06,
+      "loss": 0.0013,
+      "reward": 0.8808333277702332,
+      "reward_std": 0.07401393819600344,
+      "rewards/judge_tool_use": 0.9291666746139526,
+      "rewards/judge_tool_use/std": 0.12290690839290619,
+      "rewards/verify_correctness": 0.6875,
+      "rewards/verify_correctness/std": 0.44007769227027893,
+      "step": 480
+    },
+    {
+      "clip_ratio": 0.0029933893121778963,
+      "completion_length": 100.36111450195312,
+      "epoch": 0.97,
+      "grad_norm": 0.1513677218763824,
+      "kl": 0.52734375,
+      "learning_rate": 1e-06,
+      "loss": 0.0008,
+      "reward": 0.9199999968210856,
+      "reward_std": 0.05473165214061737,
+      "rewards/judge_tool_use": 0.9555555582046509,
+      "rewards/judge_tool_use/std": 0.0996720480422179,
+      "rewards/verify_correctness": 0.7777777711550394,
+      "rewards/verify_correctness/std": 0.3172461787859599,
+      "step": 485
+    },
+    {
+      "clip_ratio": 0.00557863749563694,
+      "completion_length": 111.61458587646484,
+      "epoch": 0.98,
+      "grad_norm": 1.166430172051773,
+      "kl": 0.385546875,
+      "learning_rate": 1e-06,
+      "loss": -0.0025,
+      "reward": 0.7954167127609253,
+      "reward_std": 0.1558975875377655,
+      "rewards/judge_tool_use": 0.8562500476837158,
+      "rewards/judge_tool_use/std": 0.24550338834524155,
+      "rewards/verify_correctness": 0.5520833283662796,
+      "rewards/verify_correctness/std": 0.46045805513858795,
+      "step": 490
+    },
+    {
+      "clip_ratio": 0.0024958730675280092,
+      "completion_length": 118.20833841959636,
+      "epoch": 0.99,
+      "grad_norm": 0.39020950446372304,
+      "kl": 0.2814453125,
+      "learning_rate": 1e-06,
+      "loss": 0.0019,
+      "reward": 0.7777777910232544,
+      "reward_std": 0.0677791138490041,
+      "rewards/judge_tool_use": 0.8055555621782938,
+      "rewards/judge_tool_use/std": 0.2596944694717725,
+      "rewards/verify_correctness": 0.6666666666666666,
+      "rewards/verify_correctness/std": 0.4650210638840993,
+      "step": 495
+    },
+    {
+      "clip_ratio": 0.004371361434459686,
+      "completion_length": 109.94791793823242,
+      "epoch": 1.0,
+      "grad_norm": 1.0635984322785825,
+      "kl": 0.35078125,
+      "learning_rate": 1e-06,
+      "loss": 0.0043,
+      "reward": 0.8491666913032532,
+      "reward_std": 0.08863747119903564,
+      "rewards/judge_tool_use": 0.8895833194255829,
+      "rewards/judge_tool_use/std": 0.23481010645627975,
+      "rewards/verify_correctness": 0.6875,
+      "rewards/verify_correctness/std": 0.46786636114120483,
+      "step": 500
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 500,
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f56fe06485fb640c762b145cf68e881acafe3ab4abf1efa8c194d9d565cc3063
+size 7672