Initial 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 +1,29 @@
+{
+  "_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
+}

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"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6079bd27eb9298907dab7fc1d166799d630eb11bda56288cc92458eab3e8c9b5
+size 988097824

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d0caaf3f9e957498d461b3ed1ff6f815e90cd4a1acb8259109b03ef511ea586
+size 1004201594

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f8e2ab643ff24c4ece7fa9a20797cdd01676f4ac256543d0994d6804130f174
+size 14180

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b4f4158f4aa521674283d4b8b46f8ac1f00ed4f557bfc5666c5b0f7b81fa1d1
+size 1064

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|>"
+}

tokenizer.json

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

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": 512,
+  "pad_token": "<|im_end|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,1333 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.21401819154628143,
+  "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": 167.21875,
+      "epoch": 0.002140181915462814,
+      "grad_norm": 3.21875,
+      "kl": 0.0,
+      "learning_rate": 2.1276595744680852e-07,
+      "loss": -0.0,
+      "reward": 0.671875,
+      "reward_std": 0.7394911348819733,
+      "rewards/correctness_reward_func": 0.28125,
+      "rewards/format_reward_func": 0.390625,
+      "step": 1
+    },
+    {
+      "completion_length": 151.84375,
+      "epoch": 0.004280363830925628,
+      "grad_norm": 3.109375,
+      "kl": 0.0,
+      "learning_rate": 4.2553191489361704e-07,
+      "loss": -0.0,
+      "reward": 0.84375,
+      "reward_std": 0.6551200225949287,
+      "rewards/correctness_reward_func": 0.34375,
+      "rewards/format_reward_func": 0.5,
+      "step": 2
+    },
+    {
+      "completion_length": 152.3125,
+      "epoch": 0.006420545746388443,
+      "grad_norm": 2.734375,
+      "kl": 0.0007652118656551465,
+      "learning_rate": 6.382978723404255e-07,
+      "loss": 0.0,
+      "reward": 0.734375,
+      "reward_std": 0.6524592041969299,
+      "rewards/correctness_reward_func": 0.21875,
+      "rewards/format_reward_func": 0.515625,
+      "step": 3
+    },
+    {
+      "completion_length": 175.328125,
+      "epoch": 0.008560727661851257,
+      "grad_norm": 2.84375,
+      "kl": 0.0007007253007031977,
+      "learning_rate": 8.510638297872341e-07,
+      "loss": 0.0,
+      "reward": 0.84375,
+      "reward_std": 0.7802355140447617,
+      "rewards/correctness_reward_func": 0.40625,
+      "rewards/format_reward_func": 0.4375,
+      "step": 4
+    },
+    {
+      "completion_length": 152.84375,
+      "epoch": 0.010700909577314071,
+      "grad_norm": 3.0625,
+      "kl": 0.0006465155893238261,
+      "learning_rate": 1.0638297872340427e-06,
+      "loss": 0.0,
+      "reward": 0.875,
+      "reward_std": 0.6669624000787735,
+      "rewards/correctness_reward_func": 0.3125,
+      "rewards/format_reward_func": 0.5625,
+      "step": 5
+    },
+    {
+      "completion_length": 179.5,
+      "epoch": 0.012841091492776886,
+      "grad_norm": 3.15625,
+      "kl": 0.0007112839666660875,
+      "learning_rate": 1.276595744680851e-06,
+      "loss": 0.0,
+      "reward": 0.6875,
+      "reward_std": 0.7003459334373474,
+      "rewards/correctness_reward_func": 0.25,
+      "rewards/format_reward_func": 0.4375,
+      "step": 6
+    },
+    {
+      "completion_length": 169.140625,
+      "epoch": 0.0149812734082397,
+      "grad_norm": 3.5,
+      "kl": 0.000702101708156988,
+      "learning_rate": 1.4893617021276596e-06,
+      "loss": 0.0,
+      "reward": 0.84375,
+      "reward_std": 0.7666550725698471,
+      "rewards/correctness_reward_func": 0.34375,
+      "rewards/format_reward_func": 0.5,
+      "step": 7
+    },
+    {
+      "completion_length": 146.390625,
+      "epoch": 0.017121455323702513,
+      "grad_norm": 3.25,
+      "kl": 0.00070322556712199,
+      "learning_rate": 1.7021276595744682e-06,
+      "loss": 0.0,
+      "reward": 0.765625,
+      "reward_std": 0.514876589179039,
+      "rewards/correctness_reward_func": 0.15625,
+      "rewards/format_reward_func": 0.609375,
+      "step": 8
+    },
+    {
+      "completion_length": 141.40625,
+      "epoch": 0.019261637239165328,
+      "grad_norm": 2.875,
+      "kl": 0.0008593319798819721,
+      "learning_rate": 1.9148936170212767e-06,
+      "loss": 0.0,
+      "reward": 1.03125,
+      "reward_std": 0.7832296937704086,
+      "rewards/correctness_reward_func": 0.46875,
+      "rewards/format_reward_func": 0.5625,
+      "step": 9
+    },
+    {
+      "completion_length": 152.21875,
+      "epoch": 0.021401819154628143,
+      "grad_norm": 3.71875,
+      "kl": 0.0011522448767209426,
+      "learning_rate": 2.1276595744680853e-06,
+      "loss": 0.0,
+      "reward": 0.75,
+      "reward_std": 0.7914351224899292,
+      "rewards/correctness_reward_func": 0.1875,
+      "rewards/format_reward_func": 0.5625,
+      "step": 10
+    },
+    {
+      "completion_length": 118.953125,
+      "epoch": 0.023542001070090957,
+      "grad_norm": 2.75,
+      "kl": 0.0009655868489062414,
+      "learning_rate": 2.340425531914894e-06,
+      "loss": 0.0,
+      "reward": 1.3125,
+      "reward_std": 0.6371002793312073,
+      "rewards/correctness_reward_func": 0.53125,
+      "rewards/format_reward_func": 0.78125,
+      "step": 11
+    },
+    {
+      "completion_length": 135.90625,
+      "epoch": 0.025682182985553772,
+      "grad_norm": 3.3125,
+      "kl": 0.0013183261326048523,
+      "learning_rate": 2.553191489361702e-06,
+      "loss": 0.0,
+      "reward": 0.890625,
+      "reward_std": 0.6318541243672371,
+      "rewards/correctness_reward_func": 0.25,
+      "rewards/format_reward_func": 0.640625,
+      "step": 12
+    },
+    {
+      "completion_length": 141.265625,
+      "epoch": 0.027822364901016586,
+      "grad_norm": 2.390625,
+      "kl": 0.0016906778037082404,
+      "learning_rate": 2.765957446808511e-06,
+      "loss": 0.0,
+      "reward": 1.015625,
+      "reward_std": 0.6712641417980194,
+      "rewards/correctness_reward_func": 0.25,
+      "rewards/format_reward_func": 0.765625,
+      "step": 13
+    },
+    {
+      "completion_length": 140.015625,
+      "epoch": 0.0299625468164794,
+      "grad_norm": 2.796875,
+      "kl": 0.0024730846926104277,
+      "learning_rate": 2.978723404255319e-06,
+      "loss": 0.0,
+      "reward": 1.0625,
+      "reward_std": 0.657661646604538,
+      "rewards/correctness_reward_func": 0.3125,
+      "rewards/format_reward_func": 0.75,
+      "step": 14
+    },
+    {
+      "completion_length": 139.65625,
+      "epoch": 0.03210272873194221,
+      "grad_norm": 2.84375,
+      "kl": 0.003585253667552024,
+      "learning_rate": 3.191489361702128e-06,
+      "loss": 0.0,
+      "reward": 1.109375,
+      "reward_std": 0.6227563470602036,
+      "rewards/correctness_reward_func": 0.3125,
+      "rewards/format_reward_func": 0.796875,
+      "step": 15
+    },
+    {
+      "completion_length": 140.671875,
+      "epoch": 0.03424291064740503,
+      "grad_norm": 2.734375,
+      "kl": 0.004379586665891111,
+      "learning_rate": 3.4042553191489363e-06,
+      "loss": 0.0,
+      "reward": 1.0,
+      "reward_std": 0.603929728269577,
+      "rewards/correctness_reward_func": 0.25,
+      "rewards/format_reward_func": 0.75,
+      "step": 16
+    },
+    {
+      "completion_length": 118.703125,
+      "epoch": 0.03638309256286784,
+      "grad_norm": 3.25,
+      "kl": 0.00417952478164807,
+      "learning_rate": 3.6170212765957453e-06,
+      "loss": 0.0,
+      "reward": 1.1875,
+      "reward_std": 0.6465111523866653,
+      "rewards/correctness_reward_func": 0.34375,
+      "rewards/format_reward_func": 0.84375,
+      "step": 17
+    },
+    {
+      "completion_length": 120.28125,
+      "epoch": 0.038523274478330656,
+      "grad_norm": 2.484375,
+      "kl": 0.005598044954240322,
+      "learning_rate": 3.8297872340425535e-06,
+      "loss": 0.0,
+      "reward": 1.09375,
+      "reward_std": 0.5800373703241348,
+      "rewards/correctness_reward_func": 0.25,
+      "rewards/format_reward_func": 0.84375,
+      "step": 18
+    },
+    {
+      "completion_length": 127.890625,
+      "epoch": 0.04066345639379347,
+      "grad_norm": 3.0,
+      "kl": 0.005727470270358026,
+      "learning_rate": 4.042553191489362e-06,
+      "loss": 0.0,
+      "reward": 1.15625,
+      "reward_std": 0.6766257882118225,
+      "rewards/correctness_reward_func": 0.34375,
+      "rewards/format_reward_func": 0.8125,
+      "step": 19
+    },
+    {
+      "completion_length": 121.703125,
+      "epoch": 0.042803638309256285,
+      "grad_norm": 2.109375,
+      "kl": 0.005017031449824572,
+      "learning_rate": 4.255319148936171e-06,
+      "loss": 0.0,
+      "reward": 1.203125,
+      "reward_std": 0.38406607508659363,
+      "rewards/correctness_reward_func": 0.28125,
+      "rewards/format_reward_func": 0.921875,
+      "step": 20
+    },
+    {
+      "completion_length": 147.4375,
+      "epoch": 0.0449438202247191,
+      "grad_norm": 2.171875,
+      "kl": 0.0043895336566492915,
+      "learning_rate": 4.468085106382979e-06,
+      "loss": 0.0,
+      "reward": 1.203125,
+      "reward_std": 0.568247452378273,
+      "rewards/correctness_reward_func": 0.34375,
+      "rewards/format_reward_func": 0.859375,
+      "step": 21
+    },
+    {
+      "completion_length": 129.40625,
+      "epoch": 0.047084002140181914,
+      "grad_norm": 1.984375,
+      "kl": 0.006280150264501572,
+      "learning_rate": 4.680851063829788e-06,
+      "loss": 0.0,
+      "reward": 1.3125,
+      "reward_std": 0.5328208804130554,
+      "rewards/correctness_reward_func": 0.34375,
+      "rewards/format_reward_func": 0.96875,
+      "step": 22
+    },
+    {
+      "completion_length": 115.859375,
+      "epoch": 0.04922418405564473,
+      "grad_norm": 3.03125,
+      "kl": 0.004840813227929175,
+      "learning_rate": 4.893617021276596e-06,
+      "loss": 0.0,
+      "reward": 1.21875,
+      "reward_std": 0.555421955883503,
+      "rewards/correctness_reward_func": 0.28125,
+      "rewards/format_reward_func": 0.9375,
+      "step": 23
+    },
+    {
+      "completion_length": 136.6875,
+      "epoch": 0.051364365971107544,
+      "grad_norm": 2.515625,
+      "kl": 0.005280128796584904,
+      "learning_rate": 5.106382978723404e-06,
+      "loss": 0.0,
+      "reward": 1.265625,
+      "reward_std": 0.6034187823534012,
+      "rewards/correctness_reward_func": 0.34375,
+      "rewards/format_reward_func": 0.921875,
+      "step": 24
+    },
+    {
+      "completion_length": 128.9375,
+      "epoch": 0.05350454788657036,
+      "grad_norm": 2.453125,
+      "kl": 0.008012514561414719,
+      "learning_rate": 5.319148936170213e-06,
+      "loss": 0.0,
+      "reward": 1.71875,
+      "reward_std": 0.9615881741046906,
+      "rewards/correctness_reward_func": 0.8125,
+      "rewards/format_reward_func": 0.90625,
+      "step": 25
+    },
+    {
+      "completion_length": 119.9375,
+      "epoch": 0.05564472980203317,
+      "grad_norm": 2.53125,
+      "kl": 0.006216021254658699,
+      "learning_rate": 5.531914893617022e-06,
+      "loss": 0.0,
+      "reward": 1.53125,
+      "reward_std": 0.6540063470602036,
+      "rewards/correctness_reward_func": 0.5625,
+      "rewards/format_reward_func": 0.96875,
+      "step": 26
+    },
+    {
+      "completion_length": 143.375,
+      "epoch": 0.05778491171749599,
+      "grad_norm": 2.5625,
+      "kl": 0.006388432695530355,
+      "learning_rate": 5.744680851063831e-06,
+      "loss": 0.0,
+      "reward": 1.296875,
+      "reward_std": 0.6433176919817924,
+      "rewards/correctness_reward_func": 0.40625,
+      "rewards/format_reward_func": 0.890625,
+      "step": 27
+    },
+    {
+      "completion_length": 131.15625,
+      "epoch": 0.0599250936329588,
+      "grad_norm": 2.453125,
+      "kl": 0.008788998704403639,
+      "learning_rate": 5.957446808510638e-06,
+      "loss": 0.0,
+      "reward": 1.546875,
+      "reward_std": 0.6077659651637077,
+      "rewards/correctness_reward_func": 0.59375,
+      "rewards/format_reward_func": 0.953125,
+      "step": 28
+    },
+    {
+      "completion_length": 124.25,
+      "epoch": 0.06206527554842162,
+      "grad_norm": 1.890625,
+      "kl": 0.0065922129433602095,
+      "learning_rate": 6.170212765957447e-06,
+      "loss": 0.0,
+      "reward": 1.4375,
+      "reward_std": 0.6141257882118225,
+      "rewards/correctness_reward_func": 0.46875,
+      "rewards/format_reward_func": 0.96875,
+      "step": 29
+    },
+    {
+      "completion_length": 128.421875,
+      "epoch": 0.06420545746388442,
+      "grad_norm": 2.203125,
+      "kl": 0.008303154725581408,
+      "learning_rate": 6.382978723404256e-06,
+      "loss": 0.0,
+      "reward": 1.625,
+      "reward_std": 0.7354267686605453,
+      "rewards/correctness_reward_func": 0.65625,
+      "rewards/format_reward_func": 0.96875,
+      "step": 30
+    },
+    {
+      "completion_length": 130.25,
+      "epoch": 0.06634563937934725,
+      "grad_norm": 1.859375,
+      "kl": 0.01200802018865943,
+      "learning_rate": 6.595744680851064e-06,
+      "loss": 0.0001,
+      "reward": 1.703125,
+      "reward_std": 0.5409187823534012,
+      "rewards/correctness_reward_func": 0.71875,
+      "rewards/format_reward_func": 0.984375,
+      "step": 31
+    },
+    {
+      "completion_length": 134.3125,
+      "epoch": 0.06848582129481005,
+      "grad_norm": 2.4375,
+      "kl": 0.012481397716328502,
+      "learning_rate": 6.808510638297873e-06,
+      "loss": 0.0001,
+      "reward": 1.5625,
+      "reward_std": 0.5936799347400665,
+      "rewards/correctness_reward_func": 0.625,
+      "rewards/format_reward_func": 0.9375,
+      "step": 32
+    },
+    {
+      "completion_length": 140.578125,
+      "epoch": 0.07062600321027288,
+      "grad_norm": 2.203125,
+      "kl": 0.011200629058293998,
+      "learning_rate": 7.021276595744682e-06,
+      "loss": 0.0001,
+      "reward": 1.46875,
+      "reward_std": 0.5570628941059113,
+      "rewards/correctness_reward_func": 0.5,
+      "rewards/format_reward_func": 0.96875,
+      "step": 33
+    },
+    {
+      "completion_length": 126.640625,
+      "epoch": 0.07276618512573568,
+      "grad_norm": 2.328125,
+      "kl": 0.011981034418568015,
+      "learning_rate": 7.234042553191491e-06,
+      "loss": 0.0001,
+      "reward": 1.671875,
+      "reward_std": 0.6420939117670059,
+      "rewards/correctness_reward_func": 0.6875,
+      "rewards/format_reward_func": 0.984375,
+      "step": 34
+    },
+    {
+      "completion_length": 141.046875,
+      "epoch": 0.0749063670411985,
+      "grad_norm": 2.0,
+      "kl": 0.01834311173297465,
+      "learning_rate": 7.446808510638298e-06,
+      "loss": 0.0001,
+      "reward": 1.75,
+      "reward_std": 0.5,
+      "rewards/correctness_reward_func": 0.75,
+      "rewards/format_reward_func": 1.0,
+      "step": 35
+    },
+    {
+      "completion_length": 129.703125,
+      "epoch": 0.07704654895666131,
+      "grad_norm": 2.15625,
+      "kl": 0.02186211384832859,
+      "learning_rate": 7.659574468085107e-06,
+      "loss": 0.0001,
+      "reward": 1.859375,
+      "reward_std": 0.6324251294136047,
+      "rewards/correctness_reward_func": 0.875,
+      "rewards/format_reward_func": 0.984375,
+      "step": 36
+    },
+    {
+      "completion_length": 147.5,
+      "epoch": 0.07918673087212413,
+      "grad_norm": 2.109375,
+      "kl": 0.018696403596550226,
+      "learning_rate": 7.872340425531916e-06,
+      "loss": 0.0001,
+      "reward": 1.78125,
+      "reward_std": 0.6237945631146431,
+      "rewards/correctness_reward_func": 0.875,
+      "rewards/format_reward_func": 0.90625,
+      "step": 37
+    },
+    {
+      "completion_length": 138.984375,
+      "epoch": 0.08132691278758694,
+      "grad_norm": 2.375,
+      "kl": 0.02475315798074007,
+      "learning_rate": 8.085106382978723e-06,
+      "loss": 0.0001,
+      "reward": 1.78125,
+      "reward_std": 0.5818375647068024,
+      "rewards/correctness_reward_func": 0.78125,
+      "rewards/format_reward_func": 1.0,
+      "step": 38
+    },
+    {
+      "completion_length": 138.875,
+      "epoch": 0.08346709470304976,
+      "grad_norm": 2.21875,
+      "kl": 0.02588808024302125,
+      "learning_rate": 8.297872340425532e-06,
+      "loss": 0.0001,
+      "reward": 1.53125,
+      "reward_std": 0.46814728528261185,
+      "rewards/correctness_reward_func": 0.59375,
+      "rewards/format_reward_func": 0.9375,
+      "step": 39
+    },
+    {
+      "completion_length": 132.46875,
+      "epoch": 0.08560727661851257,
+      "grad_norm": 2.5,
+      "kl": 0.03315168898552656,
+      "learning_rate": 8.510638297872341e-06,
+      "loss": 0.0002,
+      "reward": 1.609375,
+      "reward_std": 0.6949251294136047,
+      "rewards/correctness_reward_func": 0.625,
+      "rewards/format_reward_func": 0.984375,
+      "step": 40
+    },
+    {
+      "completion_length": 143.25,
+      "epoch": 0.08774745853397539,
+      "grad_norm": 2.34375,
+      "kl": 0.027149769477546215,
+      "learning_rate": 8.72340425531915e-06,
+      "loss": 0.0001,
+      "reward": 1.75,
+      "reward_std": 0.6540063470602036,
+      "rewards/correctness_reward_func": 0.78125,
+      "rewards/format_reward_func": 0.96875,
+      "step": 41
+    },
+    {
+      "completion_length": 149.1875,
+      "epoch": 0.0898876404494382,
+      "grad_norm": 2.734375,
+      "kl": 0.033755607437342405,
+      "learning_rate": 8.936170212765958e-06,
+      "loss": 0.0002,
+      "reward": 1.890625,
+      "reward_std": 0.9243821352720261,
+      "rewards/correctness_reward_func": 0.9375,
+      "rewards/format_reward_func": 0.953125,
+      "step": 42
+    },
+    {
+      "completion_length": 140.765625,
+      "epoch": 0.09202782236490102,
+      "grad_norm": 1.65625,
+      "kl": 0.03690116386860609,
+      "learning_rate": 9.148936170212767e-06,
+      "loss": 0.0002,
+      "reward": 1.71875,
+      "reward_std": 0.33183756470680237,
+      "rewards/correctness_reward_func": 0.71875,
+      "rewards/format_reward_func": 1.0,
+      "step": 43
+    },
+    {
+      "completion_length": 119.640625,
+      "epoch": 0.09416800428036383,
+      "grad_norm": 2.984375,
+      "kl": 0.042608937714248896,
+      "learning_rate": 9.361702127659576e-06,
+      "loss": 0.0002,
+      "reward": 2.0,
+      "reward_std": 0.6443375647068024,
+      "rewards/correctness_reward_func": 1.0,
+      "rewards/format_reward_func": 1.0,
+      "step": 44
+    },
+    {
+      "completion_length": 133.3125,
+      "epoch": 0.09630818619582665,
+      "grad_norm": 2.5625,
+      "kl": 0.05142616247758269,
+      "learning_rate": 9.574468085106385e-06,
+      "loss": 0.0003,
+      "reward": 1.5625,
+      "reward_std": 0.6292316764593124,
+      "rewards/correctness_reward_func": 0.59375,
+      "rewards/format_reward_func": 0.96875,
+      "step": 45
+    },
+    {
+      "completion_length": 137.921875,
+      "epoch": 0.09844836811128946,
+      "grad_norm": 4.65625,
+      "kl": 0.05740292742848396,
+      "learning_rate": 9.787234042553192e-06,
+      "loss": 0.0003,
+      "reward": 1.5625,
+      "reward_std": 0.39597851037979126,
+      "rewards/correctness_reward_func": 0.65625,
+      "rewards/format_reward_func": 0.90625,
+      "step": 46
+    },
+    {
+      "completion_length": 139.5625,
+      "epoch": 0.10058855002675228,
+      "grad_norm": 2.96875,
+      "kl": 0.05779360141605139,
+      "learning_rate": 1e-05,
+      "loss": 0.0003,
+      "reward": 1.75,
+      "reward_std": 0.6911139190196991,
+      "rewards/correctness_reward_func": 0.96875,
+      "rewards/format_reward_func": 0.78125,
+      "step": 47
+    },
+    {
+      "completion_length": 162.4375,
+      "epoch": 0.10272873194221509,
+      "grad_norm": 2.75,
+      "kl": 0.057580760680139065,
+      "learning_rate": 9.99986012530635e-06,
+      "loss": 0.0003,
+      "reward": 1.734375,
+      "reward_std": 0.6755875647068024,
+      "rewards/correctness_reward_func": 0.75,
+      "rewards/format_reward_func": 0.984375,
+      "step": 48
+    },
+    {
+      "completion_length": 117.859375,
+      "epoch": 0.10486891385767791,
+      "grad_norm": 1.625,
+      "kl": 0.05675748083740473,
+      "learning_rate": 9.999440509051367e-06,
+      "loss": 0.0003,
+      "reward": 2.03125,
+      "reward_std": 0.30542195588350296,
+      "rewards/correctness_reward_func": 1.0625,
+      "rewards/format_reward_func": 0.96875,
+      "step": 49
+    },
+    {
+      "completion_length": 112.28125,
+      "epoch": 0.10700909577314072,
+      "grad_norm": 3.203125,
+      "kl": 0.05932612717151642,
+      "learning_rate": 9.998741174712534e-06,
+      "loss": 0.0003,
+      "reward": 1.96875,
+      "reward_std": 0.5953208804130554,
+      "rewards/correctness_reward_func": 1.0625,
+      "rewards/format_reward_func": 0.90625,
+      "step": 50
+    },
+    {
+      "completion_length": 111.375,
+      "epoch": 0.10914927768860354,
+      "grad_norm": 2.75,
+      "kl": 0.08433620352298021,
+      "learning_rate": 9.997762161417517e-06,
+      "loss": 0.0004,
+      "reward": 1.953125,
+      "reward_std": 0.5074251294136047,
+      "rewards/correctness_reward_func": 0.96875,
+      "rewards/format_reward_func": 0.984375,
+      "step": 51
+    },
+    {
+      "completion_length": 123.78125,
+      "epoch": 0.11128945960406635,
+      "grad_norm": 3.015625,
+      "kl": 0.06248955149203539,
+      "learning_rate": 9.996503523941994e-06,
+      "loss": 0.0003,
+      "reward": 2.125,
+      "reward_std": 0.4858439266681671,
+      "rewards/correctness_reward_func": 1.125,
+      "rewards/format_reward_func": 1.0,
+      "step": 52
+    },
+    {
+      "completion_length": 119.46875,
+      "epoch": 0.11342964151952915,
+      "grad_norm": 2.625,
+      "kl": 0.061760940589010715,
+      "learning_rate": 9.994965332706574e-06,
+      "loss": 0.0003,
+      "reward": 2.0625,
+      "reward_std": 0.5,
+      "rewards/correctness_reward_func": 1.0625,
+      "rewards/format_reward_func": 1.0,
+      "step": 53
+    },
+    {
+      "completion_length": 127.265625,
+      "epoch": 0.11556982343499198,
+      "grad_norm": 1.96875,
+      "kl": 0.05814269371330738,
+      "learning_rate": 9.993147673772869e-06,
+      "loss": 0.0003,
+      "reward": 1.90625,
+      "reward_std": 0.3125,
+      "rewards/correctness_reward_func": 0.90625,
+      "rewards/format_reward_func": 1.0,
+      "step": 54
+    },
+    {
+      "completion_length": 125.84375,
+      "epoch": 0.11771000535045478,
+      "grad_norm": 2.875,
+      "kl": 0.08824395015835762,
+      "learning_rate": 9.991050648838676e-06,
+      "loss": 0.0004,
+      "reward": 2.03125,
+      "reward_std": 0.7261751294136047,
+      "rewards/correctness_reward_func": 1.03125,
+      "rewards/format_reward_func": 1.0,
+      "step": 55
+    },
+    {
+      "completion_length": 121.390625,
+      "epoch": 0.1198501872659176,
+      "grad_norm": 2.796875,
+      "kl": 0.06532257050275803,
+      "learning_rate": 9.98867437523228e-06,
+      "loss": 0.0003,
+      "reward": 1.625,
+      "reward_std": 0.46650634706020355,
+      "rewards/correctness_reward_func": 0.625,
+      "rewards/format_reward_func": 1.0,
+      "step": 56
+    },
+    {
+      "completion_length": 129.5625,
+      "epoch": 0.12199036918138041,
+      "grad_norm": 3.328125,
+      "kl": 0.06475972291082144,
+      "learning_rate": 9.986018985905901e-06,
+      "loss": 0.0003,
+      "reward": 2.015625,
+      "reward_std": 0.5409187823534012,
+      "rewards/correctness_reward_func": 1.03125,
+      "rewards/format_reward_func": 0.984375,
+      "step": 57
+    },
+    {
+      "completion_length": 133.15625,
+      "epoch": 0.12413055109684323,
+      "grad_norm": 2.0625,
+      "kl": 0.05745814461261034,
+      "learning_rate": 9.983084629428244e-06,
+      "loss": 0.0003,
+      "reward": 1.984375,
+      "reward_std": 0.42558756470680237,
+      "rewards/correctness_reward_func": 1.0,
+      "rewards/format_reward_func": 0.984375,
+      "step": 58
+    },
+    {
+      "completion_length": 124.671875,
+      "epoch": 0.12627073301230604,
+      "grad_norm": 2.9375,
+      "kl": 0.07612972147762775,
+      "learning_rate": 9.979871469976197e-06,
+      "loss": 0.0004,
+      "reward": 1.59375,
+      "reward_std": 0.6346687823534012,
+      "rewards/correctness_reward_func": 0.59375,
+      "rewards/format_reward_func": 1.0,
+      "step": 59
+    },
+    {
+      "completion_length": 109.921875,
+      "epoch": 0.12841091492776885,
+      "grad_norm": 3.0,
+      "kl": 0.06931339204311371,
+      "learning_rate": 9.976379687325633e-06,
+      "loss": 0.0003,
+      "reward": 1.96875,
+      "reward_std": 0.6540063470602036,
+      "rewards/correctness_reward_func": 0.96875,
+      "rewards/format_reward_func": 1.0,
+      "step": 60
+    },
+    {
+      "completion_length": 125.53125,
+      "epoch": 0.13055109684323168,
+      "grad_norm": 2.546875,
+      "kl": 0.05608335882425308,
+      "learning_rate": 9.972609476841368e-06,
+      "loss": 0.0003,
+      "reward": 2.125,
+      "reward_std": 0.5915063470602036,
+      "rewards/correctness_reward_func": 1.125,
+      "rewards/format_reward_func": 1.0,
+      "step": 61
+    },
+    {
+      "completion_length": 127.8125,
+      "epoch": 0.1326912787586945,
+      "grad_norm": 2.59375,
+      "kl": 0.14377161115407944,
+      "learning_rate": 9.968561049466214e-06,
+      "loss": 0.0007,
+      "reward": 1.84375,
+      "reward_std": 0.5290063321590424,
+      "rewards/correctness_reward_func": 0.84375,
+      "rewards/format_reward_func": 1.0,
+      "step": 62
+    },
+    {
+      "completion_length": 124.75,
+      "epoch": 0.1348314606741573,
+      "grad_norm": 1.6328125,
+      "kl": 0.06179473642259836,
+      "learning_rate": 9.964234631709188e-06,
+      "loss": 0.0003,
+      "reward": 1.90625,
+      "reward_std": 0.2596687823534012,
+      "rewards/correctness_reward_func": 0.90625,
+      "rewards/format_reward_func": 1.0,
+      "step": 63
+    },
+    {
+      "completion_length": 125.515625,
+      "epoch": 0.1369716425896201,
+      "grad_norm": 2.875,
+      "kl": 0.08027161657810211,
+      "learning_rate": 9.959630465632833e-06,
+      "loss": 0.0004,
+      "reward": 2.15625,
+      "reward_std": 0.6011751294136047,
+      "rewards/correctness_reward_func": 1.15625,
+      "rewards/format_reward_func": 1.0,
+      "step": 64
+    },
+    {
+      "completion_length": 114.890625,
+      "epoch": 0.13911182450508294,
+      "grad_norm": 2.90625,
+      "kl": 0.13545920699834824,
+      "learning_rate": 9.954748808839675e-06,
+      "loss": 0.0007,
+      "reward": 1.859375,
+      "reward_std": 0.4826504588127136,
+      "rewards/correctness_reward_func": 0.875,
+      "rewards/format_reward_func": 0.984375,
+      "step": 65
+    },
+    {
+      "completion_length": 121.125,
+      "epoch": 0.14125200642054575,
+      "grad_norm": 1.9921875,
+      "kl": 0.07298032753169537,
+      "learning_rate": 9.949589934457815e-06,
+      "loss": 0.0004,
+      "reward": 2.03125,
+      "reward_std": 0.3846687823534012,
+      "rewards/correctness_reward_func": 1.03125,
+      "rewards/format_reward_func": 1.0,
+      "step": 66
+    },
+    {
+      "completion_length": 122.625,
+      "epoch": 0.14339218833600856,
+      "grad_norm": 3.359375,
+      "kl": 0.10324916429817677,
+      "learning_rate": 9.944154131125643e-06,
+      "loss": 0.0005,
+      "reward": 2.078125,
+      "reward_std": 0.5409187823534012,
+      "rewards/correctness_reward_func": 1.09375,
+      "rewards/format_reward_func": 0.984375,
+      "step": 67
+    },
+    {
+      "completion_length": 133.40625,
+      "epoch": 0.14553237025147137,
+      "grad_norm": 2.28125,
+      "kl": 0.10124836303293705,
+      "learning_rate": 9.938441702975689e-06,
+      "loss": 0.0005,
+      "reward": 1.71875,
+      "reward_std": 0.40400634706020355,
+      "rewards/correctness_reward_func": 0.71875,
+      "rewards/format_reward_func": 1.0,
+      "step": 68
+    },
+    {
+      "completion_length": 141.984375,
+      "epoch": 0.1476725521669342,
+      "grad_norm": 2.921875,
+      "kl": 0.08171628974378109,
+      "learning_rate": 9.932452969617607e-06,
+      "loss": 0.0004,
+      "reward": 1.96875,
+      "reward_std": 0.41325797885656357,
+      "rewards/correctness_reward_func": 1.0,
+      "rewards/format_reward_func": 0.96875,
+      "step": 69
+    },
+    {
+      "completion_length": 125.609375,
+      "epoch": 0.149812734082397,
+      "grad_norm": 2.53125,
+      "kl": 0.08117420598864555,
+      "learning_rate": 9.926188266120297e-06,
+      "loss": 0.0004,
+      "reward": 1.8125,
+      "reward_std": 0.41367512941360474,
+      "rewards/correctness_reward_func": 0.8125,
+      "rewards/format_reward_func": 1.0,
+      "step": 70
+    },
+    {
+      "completion_length": 122.296875,
+      "epoch": 0.15195291599785982,
+      "grad_norm": 2.875,
+      "kl": 0.08432869054377079,
+      "learning_rate": 9.91964794299315e-06,
+      "loss": 0.0004,
+      "reward": 2.03125,
+      "reward_std": 0.5818375647068024,
+      "rewards/correctness_reward_func": 1.03125,
+      "rewards/format_reward_func": 1.0,
+      "step": 71
+    },
+    {
+      "completion_length": 136.359375,
+      "epoch": 0.15409309791332262,
+      "grad_norm": 2.296875,
+      "kl": 0.10322786308825016,
+      "learning_rate": 9.912832366166443e-06,
+      "loss": 0.0005,
+      "reward": 1.5625,
+      "reward_std": 0.26933756470680237,
+      "rewards/correctness_reward_func": 0.5625,
+      "rewards/format_reward_func": 1.0,
+      "step": 72
+    },
+    {
+      "completion_length": 113.921875,
+      "epoch": 0.15623327982878546,
+      "grad_norm": 3.84375,
+      "kl": 0.13962376862764359,
+      "learning_rate": 9.905741916970863e-06,
+      "loss": 0.0007,
+      "reward": 2.09375,
+      "reward_std": 0.5290063470602036,
+      "rewards/correctness_reward_func": 1.09375,
+      "rewards/format_reward_func": 1.0,
+      "step": 73
+    },
+    {
+      "completion_length": 120.5625,
+      "epoch": 0.15837346174424827,
+      "grad_norm": 3.125,
+      "kl": 0.10202482901513577,
+      "learning_rate": 9.898376992116179e-06,
+      "loss": 0.0005,
+      "reward": 1.796875,
+      "reward_std": 0.5892626941204071,
+      "rewards/correctness_reward_func": 0.8125,
+      "rewards/format_reward_func": 0.984375,
+      "step": 74
+    },
+    {
+      "completion_length": 132.390625,
+      "epoch": 0.16051364365971107,
+      "grad_norm": 2.875,
+      "kl": 0.09189064055681229,
+      "learning_rate": 9.890738003669029e-06,
+      "loss": 0.0005,
+      "reward": 2.125,
+      "reward_std": 0.6636751294136047,
+      "rewards/correctness_reward_func": 1.125,
+      "rewards/format_reward_func": 1.0,
+      "step": 75
+    },
+    {
+      "completion_length": 127.546875,
+      "epoch": 0.16265382557517388,
+      "grad_norm": 3.3125,
+      "kl": 0.09288744628429413,
+      "learning_rate": 9.882825379029883e-06,
+      "loss": 0.0005,
+      "reward": 1.84375,
+      "reward_std": 0.7068375498056412,
+      "rewards/correctness_reward_func": 0.84375,
+      "rewards/format_reward_func": 1.0,
+      "step": 76
+    },
+    {
+      "completion_length": 116.90625,
+      "epoch": 0.1647940074906367,
+      "grad_norm": 2.59375,
+      "kl": 0.10366734117269516,
+      "learning_rate": 9.874639560909118e-06,
+      "loss": 0.0005,
+      "reward": 1.625,
+      "reward_std": 0.375,
+      "rewards/correctness_reward_func": 0.625,
+      "rewards/format_reward_func": 1.0,
+      "step": 77
+    },
+    {
+      "completion_length": 145.9375,
+      "epoch": 0.16693418940609953,
+      "grad_norm": 3.125,
+      "kl": 0.14030247181653976,
+      "learning_rate": 9.866181007302258e-06,
+      "loss": 0.0007,
+      "reward": 1.640625,
+      "reward_std": 0.43525634706020355,
+      "rewards/correctness_reward_func": 0.65625,
+      "rewards/format_reward_func": 0.984375,
+      "step": 78
+    },
+    {
+      "completion_length": 114.1875,
+      "epoch": 0.16907437132156233,
+      "grad_norm": 4.15625,
+      "kl": 0.10448622144758701,
+      "learning_rate": 9.857450191464337e-06,
+      "loss": 0.0005,
+      "reward": 1.96875,
+      "reward_std": 0.7790063321590424,
+      "rewards/correctness_reward_func": 0.96875,
+      "rewards/format_reward_func": 1.0,
+      "step": 79
+    },
+    {
+      "completion_length": 135.78125,
+      "epoch": 0.17121455323702514,
+      "grad_norm": 2.421875,
+      "kl": 0.08989986591041088,
+      "learning_rate": 9.848447601883436e-06,
+      "loss": 0.0004,
+      "reward": 1.40625,
+      "reward_std": 0.3582531735301018,
+      "rewards/correctness_reward_func": 0.4375,
+      "rewards/format_reward_func": 0.96875,
+      "step": 80
+    },
+    {
+      "completion_length": 141.953125,
+      "epoch": 0.17335473515248795,
+      "grad_norm": 2.953125,
+      "kl": 0.07188372500240803,
+      "learning_rate": 9.839173742253334e-06,
+      "loss": 0.0004,
+      "reward": 1.96875,
+      "reward_std": 0.6733439117670059,
+      "rewards/correctness_reward_func": 1.0,
+      "rewards/format_reward_func": 0.96875,
+      "step": 81
+    },
+    {
+      "completion_length": 118.359375,
+      "epoch": 0.17549491706795078,
+      "grad_norm": 1.7109375,
+      "kl": 0.07351219840347767,
+      "learning_rate": 9.829629131445342e-06,
+      "loss": 0.0004,
+      "reward": 2.21875,
+      "reward_std": 0.27900634706020355,
+      "rewards/correctness_reward_func": 1.21875,
+      "rewards/format_reward_func": 1.0,
+      "step": 82
+    },
+    {
+      "completion_length": 127.109375,
+      "epoch": 0.1776350989834136,
+      "grad_norm": 1.359375,
+      "kl": 0.07365736179053783,
+      "learning_rate": 9.819814303479268e-06,
+      "loss": 0.0004,
+      "reward": 2.1875,
+      "reward_std": 0.26933756470680237,
+      "rewards/correctness_reward_func": 1.1875,
+      "rewards/format_reward_func": 1.0,
+      "step": 83
+    },
+    {
+      "completion_length": 139.078125,
+      "epoch": 0.1797752808988764,
+      "grad_norm": 2.484375,
+      "kl": 0.07714031636714935,
+      "learning_rate": 9.80972980749353e-06,
+      "loss": 0.0004,
+      "reward": 1.90625,
+      "reward_std": 0.6540063470602036,
+      "rewards/correctness_reward_func": 0.90625,
+      "rewards/format_reward_func": 1.0,
+      "step": 84
+    },
+    {
+      "completion_length": 128.390625,
+      "epoch": 0.1819154628143392,
+      "grad_norm": 1.4453125,
+      "kl": 0.056750121526420116,
+      "learning_rate": 9.799376207714446e-06,
+      "loss": 0.0003,
+      "reward": 2.125,
+      "reward_std": 0.25,
+      "rewards/correctness_reward_func": 1.15625,
+      "rewards/format_reward_func": 0.96875,
+      "step": 85
+    },
+    {
+      "completion_length": 160.234375,
+      "epoch": 0.18405564472980204,
+      "grad_norm": 1.9296875,
+      "kl": 0.06542882043868303,
+      "learning_rate": 9.788754083424654e-06,
+      "loss": 0.0003,
+      "reward": 1.734375,
+      "reward_std": 0.42558756470680237,
+      "rewards/correctness_reward_func": 0.75,
+      "rewards/format_reward_func": 0.984375,
+      "step": 86
+    },
+    {
+      "completion_length": 120.546875,
+      "epoch": 0.18619582664526485,
+      "grad_norm": 2.703125,
+      "kl": 0.14421744085848331,
+      "learning_rate": 9.777864028930705e-06,
+      "loss": 0.0007,
+      "reward": 2.140625,
+      "reward_std": 0.6324251294136047,
+      "rewards/correctness_reward_func": 1.15625,
+      "rewards/format_reward_func": 0.984375,
+      "step": 87
+    },
+    {
+      "completion_length": 137.296875,
+      "epoch": 0.18833600856072766,
+      "grad_norm": 2.59375,
+      "kl": 0.0669497437775135,
+      "learning_rate": 9.766706653529814e-06,
+      "loss": 0.0003,
+      "reward": 2.078125,
+      "reward_std": 0.7763455435633659,
+      "rewards/correctness_reward_func": 1.09375,
+      "rewards/format_reward_func": 0.984375,
+      "step": 88
+    },
+    {
+      "completion_length": 145.4375,
+      "epoch": 0.19047619047619047,
+      "grad_norm": 2.609375,
+      "kl": 0.07120031677186489,
+      "learning_rate": 9.755282581475769e-06,
+      "loss": 0.0004,
+      "reward": 1.59375,
+      "reward_std": 0.7261751294136047,
+      "rewards/correctness_reward_func": 0.59375,
+      "rewards/format_reward_func": 1.0,
+      "step": 89
+    },
+    {
+      "completion_length": 137.59375,
+      "epoch": 0.1926163723916533,
+      "grad_norm": 2.484375,
+      "kl": 0.07096913084387779,
+      "learning_rate": 9.743592451944e-06,
+      "loss": 0.0004,
+      "reward": 1.828125,
+      "reward_std": 0.6324251145124435,
+      "rewards/correctness_reward_func": 0.84375,
+      "rewards/format_reward_func": 0.984375,
+      "step": 90
+    },
+    {
+      "completion_length": 117.359375,
+      "epoch": 0.1947565543071161,
+      "grad_norm": 2.046875,
+      "kl": 0.06865348853170872,
+      "learning_rate": 9.731636918995821e-06,
+      "loss": 0.0003,
+      "reward": 2.3125,
+      "reward_std": 0.39433756470680237,
+      "rewards/correctness_reward_func": 1.3125,
+      "rewards/format_reward_func": 1.0,
+      "step": 91
+    },
+    {
+      "completion_length": 130.734375,
+      "epoch": 0.19689673622257892,
+      "grad_norm": 2.453125,
+      "kl": 0.08213793113827705,
+      "learning_rate": 9.719416651541839e-06,
+      "loss": 0.0004,
+      "reward": 2.1875,
+      "reward_std": 0.5818375647068024,
+      "rewards/correctness_reward_func": 1.21875,
+      "rewards/format_reward_func": 0.96875,
+      "step": 92
+    },
+    {
+      "completion_length": 143.015625,
+      "epoch": 0.19903691813804172,
+      "grad_norm": 1.2578125,
+      "kl": 0.07020723912864923,
+      "learning_rate": 9.706932333304518e-06,
+      "loss": 0.0004,
+      "reward": 1.78125,
+      "reward_std": 0.2596687823534012,
+      "rewards/correctness_reward_func": 0.78125,
+      "rewards/format_reward_func": 1.0,
+      "step": 93
+    },
+    {
+      "completion_length": 119.96875,
+      "epoch": 0.20117710005350456,
+      "grad_norm": 2.375,
+      "kl": 0.07560589909553528,
+      "learning_rate": 9.694184662779931e-06,
+      "loss": 0.0004,
+      "reward": 2.125,
+      "reward_std": 0.46650633215904236,
+      "rewards/correctness_reward_func": 1.125,
+      "rewards/format_reward_func": 1.0,
+      "step": 94
+    },
+    {
+      "completion_length": 139.21875,
+      "epoch": 0.20331728196896737,
+      "grad_norm": 2.078125,
+      "kl": 0.09026694297790527,
+      "learning_rate": 9.681174353198687e-06,
+      "loss": 0.0005,
+      "reward": 2.1875,
+      "reward_std": 0.5386751443147659,
+      "rewards/correctness_reward_func": 1.1875,
+      "rewards/format_reward_func": 1.0,
+      "step": 95
+    },
+    {
+      "completion_length": 151.328125,
+      "epoch": 0.20545746388443017,
+      "grad_norm": 2.25,
+      "kl": 0.06551774125546217,
+      "learning_rate": 9.667902132486009e-06,
+      "loss": 0.0003,
+      "reward": 1.75,
+      "reward_std": 0.4471687823534012,
+      "rewards/correctness_reward_func": 0.75,
+      "rewards/format_reward_func": 1.0,
+      "step": 96
+    },
+    {
+      "completion_length": 138.640625,
+      "epoch": 0.20759764579989298,
+      "grad_norm": 2.40625,
+      "kl": 0.07146297954022884,
+      "learning_rate": 9.654368743221022e-06,
+      "loss": 0.0004,
+      "reward": 1.796875,
+      "reward_std": 0.6604816764593124,
+      "rewards/correctness_reward_func": 0.8125,
+      "rewards/format_reward_func": 0.984375,
+      "step": 97
+    },
+    {
+      "completion_length": 140.9375,
+      "epoch": 0.20973782771535582,
+      "grad_norm": 2.1875,
+      "kl": 0.09298614785075188,
+      "learning_rate": 9.640574942595195e-06,
+      "loss": 0.0005,
+      "reward": 2.09375,
+      "reward_std": 0.5818375647068024,
+      "rewards/correctness_reward_func": 1.09375,
+      "rewards/format_reward_func": 1.0,
+      "step": 98
+    },
+    {
+      "completion_length": 151.171875,
+      "epoch": 0.21187800963081863,
+      "grad_norm": 2.5625,
+      "kl": 0.06539157032966614,
+      "learning_rate": 9.626521502369984e-06,
+      "loss": 0.0003,
+      "reward": 2.3125,
+      "reward_std": 0.6443375647068024,
+      "rewards/correctness_reward_func": 1.34375,
+      "rewards/format_reward_func": 0.96875,
+      "step": 99
+    },
+    {
+      "completion_length": 154.21875,
+      "epoch": 0.21401819154628143,
+      "grad_norm": 2.171875,
+      "kl": 0.07382548321038485,
+      "learning_rate": 9.612209208833648e-06,
+      "loss": 0.0004,
+      "reward": 1.9375,
+      "reward_std": 0.6292316764593124,
+      "rewards/correctness_reward_func": 0.96875,
+      "rewards/format_reward_func": 0.96875,
+      "step": 100
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 467,
+  "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": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a891607919cc3a4f5685413eac493180dbc7f55029869807ee308abd66d3a017
+size 5752