Upload folder using huggingface_hub

.gitattributes

@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 Qwn3-0.6B-tldr-lora-f16.gguf filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-0.6B",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05246882d23ac9824d92c773ebe7776a2e466c9f28d2cddf9c201f440004123a
+size 4602248

added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e60f193253c04fa0140af875dc45941007ad54298afaefcbc30a22d2702bf345
+size 9269306

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0596da8c3c9a5924d6a76701e5038b6b92bc617904f4c98950fd4451f68ba72a
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f657ace31ad19492aa41ca6863f9765d60806fab555c109efce7ed3e984a21d2
+size 1064

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 sha256:67cc0080ffd7555f723f423c27cfef314e1ad9d335c8b79f465c5faba1ed478b
+size 11422821

tokenizer_config.json

@@ -0,0 +1,241 @@
+{
+  "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
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "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\\n' }}\n    {%- endif %}\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    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set content = message.content %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in message.content %}\n                {%- set content = message.content.split('</think>')[-1].lstrip('\\n') %}\n                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first 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    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,1354 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.009424101711759565,
+  "eval_steps": 500,
+  "global_step": 1100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 117.2125,
+      "epoch": 8.567365192508696e-05,
+      "grad_norm": 1.5428545475006104,
+      "kl": 7.6057333103563e-06,
+      "learning_rate": 9.999742979044224e-07,
+      "loss": 0.0,
+      "reward": -304.1875,
+      "reward_std": 100.79027481079102,
+      "rewards/reward_len": -304.1875,
+      "step": 10
+    },
+    {
+      "completion_length": 106.55,
+      "epoch": 0.00017134730385017391,
+      "grad_norm": 1.7018674612045288,
+      "kl": 4.6649108844576405e-05,
+      "learning_rate": 9.999457400204474e-07,
+      "loss": 0.0,
+      "reward": -268.175,
+      "reward_std": 85.13249053955079,
+      "rewards/reward_len": -268.175,
+      "step": 20
+    },
+    {
+      "completion_length": 114.1,
+      "epoch": 0.00025702095577526086,
+      "grad_norm": 1.4646943807601929,
+      "kl": 0.00014046951691852881,
+      "learning_rate": 9.999171821364723e-07,
+      "loss": 0.0,
+      "reward": -295.625,
+      "reward_std": 92.77591400146484,
+      "rewards/reward_len": -295.625,
+      "step": 30
+    },
+    {
+      "completion_length": 112.45,
+      "epoch": 0.00034269460770034783,
+      "grad_norm": 1.4721007347106934,
+      "kl": 0.0002871266377042048,
+      "learning_rate": 9.998886242524973e-07,
+      "loss": 0.0,
+      "reward": -287.4375,
+      "reward_std": 98.72801208496094,
+      "rewards/reward_len": -287.4375,
+      "step": 40
+    },
+    {
+      "completion_length": 103.3375,
+      "epoch": 0.0004283682596254348,
+      "grad_norm": 1.5164705514907837,
+      "kl": 0.00045935034868307414,
+      "learning_rate": 9.998600663685224e-07,
+      "loss": 0.0,
+      "reward": -254.0625,
+      "reward_std": 92.05970458984375,
+      "rewards/reward_len": -254.0625,
+      "step": 50
+    },
+    {
+      "completion_length": 100.4375,
+      "epoch": 0.0005140419115505217,
+      "grad_norm": 1.267320156097412,
+      "kl": 0.0006870983401313424,
+      "learning_rate": 9.998315084845474e-07,
+      "loss": 0.0,
+      "reward": -229.725,
+      "reward_std": 107.81015129089356,
+      "rewards/reward_len": -229.725,
+      "step": 60
+    },
+    {
+      "completion_length": 104.0375,
+      "epoch": 0.0005997155634756087,
+      "grad_norm": 1.2804882526397705,
+      "kl": 0.0007431535341311246,
+      "learning_rate": 9.998029506005722e-07,
+      "loss": 0.0,
+      "reward": -276.275,
+      "reward_std": 99.2921049118042,
+      "rewards/reward_len": -276.275,
+      "step": 70
+    },
+    {
+      "completion_length": 102.2875,
+      "epoch": 0.0006853892154006957,
+      "grad_norm": 1.4880214929580688,
+      "kl": 0.0008916158636566252,
+      "learning_rate": 9.997743927165973e-07,
+      "loss": 0.0,
+      "reward": -266.025,
+      "reward_std": 89.98284072875977,
+      "rewards/reward_len": -266.025,
+      "step": 80
+    },
+    {
+      "completion_length": 124.3625,
+      "epoch": 0.0007710628673257826,
+      "grad_norm": 1.6408586502075195,
+      "kl": 0.0010400827683042734,
+      "learning_rate": 9.997458348326221e-07,
+      "loss": 0.0,
+      "reward": -318.0375,
+      "reward_std": 99.00584564208984,
+      "rewards/reward_len": -318.0375,
+      "step": 90
+    },
+    {
+      "completion_length": 92.075,
+      "epoch": 0.0008567365192508696,
+      "grad_norm": 1.535640835762024,
+      "kl": 0.001465974550228566,
+      "learning_rate": 9.997172769486472e-07,
+      "loss": 0.0001,
+      "reward": -206.9125,
+      "reward_std": 85.6998176574707,
+      "rewards/reward_len": -206.9125,
+      "step": 100
+    },
+    {
+      "completion_length": 99.125,
+      "epoch": 0.0009424101711759565,
+      "grad_norm": 2.4441370964050293,
+      "kl": 0.0018776856479234994,
+      "learning_rate": 9.99688719064672e-07,
+      "loss": 0.0001,
+      "reward": -247.55,
+      "reward_std": 92.91195907592774,
+      "rewards/reward_len": -247.55,
+      "step": 110
+    },
+    {
+      "completion_length": 104.35,
+      "epoch": 0.0010280838231010434,
+      "grad_norm": 1.3777748346328735,
+      "kl": 0.002180302981287241,
+      "learning_rate": 9.99660161180697e-07,
+      "loss": 0.0001,
+      "reward": -257.2625,
+      "reward_std": 95.76971626281738,
+      "rewards/reward_len": -257.2625,
+      "step": 120
+    },
+    {
+      "completion_length": 94.4125,
+      "epoch": 0.0011137574750261306,
+      "grad_norm": 1.3118126392364502,
+      "kl": 0.002771495224442333,
+      "learning_rate": 9.99631603296722e-07,
+      "loss": 0.0001,
+      "reward": -202.2125,
+      "reward_std": 84.88922119140625,
+      "rewards/reward_len": -202.2125,
+      "step": 130
+    },
+    {
+      "completion_length": 89.275,
+      "epoch": 0.0011994311269512175,
+      "grad_norm": 1.551858901977539,
+      "kl": 0.0034235142404213546,
+      "learning_rate": 9.996030454127471e-07,
+      "loss": 0.0001,
+      "reward": -211.7375,
+      "reward_std": 80.14296417236328,
+      "rewards/reward_len": -211.7375,
+      "step": 140
+    },
+    {
+      "completion_length": 115.8,
+      "epoch": 0.0012851047788763044,
+      "grad_norm": 1.333330750465393,
+      "kl": 0.0028527263493742794,
+      "learning_rate": 9.99574487528772e-07,
+      "loss": 0.0001,
+      "reward": -279.3625,
+      "reward_std": 72.2537956237793,
+      "rewards/reward_len": -279.3625,
+      "step": 150
+    },
+    {
+      "completion_length": 113.575,
+      "epoch": 0.0013707784308013913,
+      "grad_norm": 1.7015572786331177,
+      "kl": 0.0033493154449388387,
+      "learning_rate": 9.99545929644797e-07,
+      "loss": 0.0001,
+      "reward": -274.15,
+      "reward_std": 101.25760879516602,
+      "rewards/reward_len": -274.15,
+      "step": 160
+    },
+    {
+      "completion_length": 106.65,
+      "epoch": 0.0014564520827264782,
+      "grad_norm": 1.95174241065979,
+      "kl": 0.004216998419724405,
+      "learning_rate": 9.995173717608219e-07,
+      "loss": 0.0002,
+      "reward": -263.525,
+      "reward_std": 89.78757858276367,
+      "rewards/reward_len": -263.525,
+      "step": 170
+    },
+    {
+      "completion_length": 97.4875,
+      "epoch": 0.0015421257346515652,
+      "grad_norm": 1.5441001653671265,
+      "kl": 0.0053229076787829396,
+      "learning_rate": 9.99488813876847e-07,
+      "loss": 0.0002,
+      "reward": -236.775,
+      "reward_std": 93.90848999023437,
+      "rewards/reward_len": -236.775,
+      "step": 180
+    },
+    {
+      "completion_length": 102.7125,
+      "epoch": 0.0016277993865766523,
+      "grad_norm": 1.435225009918213,
+      "kl": 0.005796084133908153,
+      "learning_rate": 9.99460255992872e-07,
+      "loss": 0.0002,
+      "reward": -243.8375,
+      "reward_std": 95.30473594665527,
+      "rewards/reward_len": -243.8375,
+      "step": 190
+    },
+    {
+      "completion_length": 112.8,
+      "epoch": 0.0017134730385017392,
+      "grad_norm": 1.3907208442687988,
+      "kl": 0.006385851232334971,
+      "learning_rate": 9.994316981088968e-07,
+      "loss": 0.0003,
+      "reward": -292.475,
+      "reward_std": 99.54066009521485,
+      "rewards/reward_len": -292.475,
+      "step": 200
+    },
+    {
+      "completion_length": 106.2875,
+      "epoch": 0.0017991466904268261,
+      "grad_norm": 1.0377769470214844,
+      "kl": 0.00682140497956425,
+      "learning_rate": 9.994031402249218e-07,
+      "loss": 0.0003,
+      "reward": -263.15,
+      "reward_std": 81.53659133911133,
+      "rewards/reward_len": -263.15,
+      "step": 210
+    },
+    {
+      "completion_length": 94.15,
+      "epoch": 0.001884820342351913,
+      "grad_norm": 1.9323745965957642,
+      "kl": 0.007357935793697834,
+      "learning_rate": 9.993745823409469e-07,
+      "loss": 0.0003,
+      "reward": -217.4625,
+      "reward_std": 90.16817970275879,
+      "rewards/reward_len": -217.4625,
+      "step": 220
+    },
+    {
+      "completion_length": 92.8,
+      "epoch": 0.001970493994277,
+      "grad_norm": 1.8935438394546509,
+      "kl": 0.008906404627487064,
+      "learning_rate": 9.993460244569717e-07,
+      "loss": 0.0004,
+      "reward": -220.675,
+      "reward_std": 74.1481372833252,
+      "rewards/reward_len": -220.675,
+      "step": 230
+    },
+    {
+      "completion_length": 103.4875,
+      "epoch": 0.002056167646202087,
+      "grad_norm": 1.4551963806152344,
+      "kl": 0.008559457911178469,
+      "learning_rate": 9.993174665729968e-07,
+      "loss": 0.0003,
+      "reward": -242.5,
+      "reward_std": 96.66918106079102,
+      "rewards/reward_len": -242.5,
+      "step": 240
+    },
+    {
+      "completion_length": 96.7375,
+      "epoch": 0.002141841298127174,
+      "grad_norm": 1.3759669065475464,
+      "kl": 0.010590891446918249,
+      "learning_rate": 9.992889086890218e-07,
+      "loss": 0.0004,
+      "reward": -240.6125,
+      "reward_std": 107.75033836364746,
+      "rewards/reward_len": -240.6125,
+      "step": 250
+    },
+    {
+      "completion_length": 95.9125,
+      "epoch": 0.002227514950052261,
+      "grad_norm": 1.7361980676651,
+      "kl": 0.009502861043438315,
+      "learning_rate": 9.992603508050467e-07,
+      "loss": 0.0004,
+      "reward": -219.2125,
+      "reward_std": 86.29145584106445,
+      "rewards/reward_len": -219.2125,
+      "step": 260
+    },
+    {
+      "completion_length": 91.35,
+      "epoch": 0.002313188601977348,
+      "grad_norm": 1.6296714544296265,
+      "kl": 0.012998782657086849,
+      "learning_rate": 9.992317929210717e-07,
+      "loss": 0.0005,
+      "reward": -205.9,
+      "reward_std": 104.51896514892579,
+      "rewards/reward_len": -205.9,
+      "step": 270
+    },
+    {
+      "completion_length": 92.825,
+      "epoch": 0.002398862253902435,
+      "grad_norm": 2.636399269104004,
+      "kl": 0.013483931683003903,
+      "learning_rate": 9.992032350370967e-07,
+      "loss": 0.0005,
+      "reward": -211.0375,
+      "reward_std": 72.49085922241211,
+      "rewards/reward_len": -211.0375,
+      "step": 280
+    },
+    {
+      "completion_length": 96.475,
+      "epoch": 0.002484535905827522,
+      "grad_norm": 1.7585633993148804,
+      "kl": 0.013156655337661504,
+      "learning_rate": 9.991746771531216e-07,
+      "loss": 0.0005,
+      "reward": -233.4,
+      "reward_std": 77.94148178100586,
+      "rewards/reward_len": -233.4,
+      "step": 290
+    },
+    {
+      "completion_length": 78.575,
+      "epoch": 0.002570209557752609,
+      "grad_norm": 1.8656781911849976,
+      "kl": 0.019020712468773127,
+      "learning_rate": 9.991461192691466e-07,
+      "loss": 0.0008,
+      "reward": -165.3625,
+      "reward_std": 83.65277671813965,
+      "rewards/reward_len": -165.3625,
+      "step": 300
+    },
+    {
+      "completion_length": 80.9,
+      "epoch": 0.0026558832096776957,
+      "grad_norm": 1.886581540107727,
+      "kl": 0.0151671776548028,
+      "learning_rate": 9.991175613851715e-07,
+      "loss": 0.0006,
+      "reward": -153.625,
+      "reward_std": 61.95796432495117,
+      "rewards/reward_len": -153.625,
+      "step": 310
+    },
+    {
+      "completion_length": 79.9,
+      "epoch": 0.0027415568616027826,
+      "grad_norm": 3.363229990005493,
+      "kl": 0.018376798275858162,
+      "learning_rate": 9.990890035011965e-07,
+      "loss": 0.0007,
+      "reward": -170.775,
+      "reward_std": 83.77405395507813,
+      "rewards/reward_len": -170.775,
+      "step": 320
+    },
+    {
+      "completion_length": 90.9625,
+      "epoch": 0.0028272305135278696,
+      "grad_norm": 3.4404351711273193,
+      "kl": 0.018607683992013336,
+      "learning_rate": 9.990604456172216e-07,
+      "loss": 0.0007,
+      "reward": -189.1625,
+      "reward_std": 93.11449737548828,
+      "rewards/reward_len": -189.1625,
+      "step": 330
+    },
+    {
+      "completion_length": 90.8375,
+      "epoch": 0.0029129041654529565,
+      "grad_norm": 1.3329404592514038,
+      "kl": 0.01699499851092696,
+      "learning_rate": 9.990318877332464e-07,
+      "loss": 0.0007,
+      "reward": -203.4,
+      "reward_std": 93.15925168991089,
+      "rewards/reward_len": -203.4,
+      "step": 340
+    },
+    {
+      "completion_length": 93.7125,
+      "epoch": 0.0029985778173780434,
+      "grad_norm": 2.384962797164917,
+      "kl": 0.022056785970926286,
+      "learning_rate": 9.990033298492714e-07,
+      "loss": 0.0009,
+      "reward": -209.9,
+      "reward_std": 95.43540840148925,
+      "rewards/reward_len": -209.9,
+      "step": 350
+    },
+    {
+      "completion_length": 77.975,
+      "epoch": 0.0030842514693031303,
+      "grad_norm": 2.1530113220214844,
+      "kl": 0.02068719444796443,
+      "learning_rate": 9.989747719652965e-07,
+      "loss": 0.0008,
+      "reward": -145.9,
+      "reward_std": 78.50229835510254,
+      "rewards/reward_len": -145.9,
+      "step": 360
+    },
+    {
+      "completion_length": 77.175,
+      "epoch": 0.0031699251212282177,
+      "grad_norm": 2.1676974296569824,
+      "kl": 0.028167350962758065,
+      "learning_rate": 9.989462140813213e-07,
+      "loss": 0.0011,
+      "reward": -152.125,
+      "reward_std": 77.3315586090088,
+      "rewards/reward_len": -152.125,
+      "step": 370
+    },
+    {
+      "completion_length": 93.9,
+      "epoch": 0.0032555987731533046,
+      "grad_norm": 2.6751251220703125,
+      "kl": 0.028882794454693793,
+      "learning_rate": 9.989176561973464e-07,
+      "loss": 0.0012,
+      "reward": -192.5875,
+      "reward_std": 93.17231407165528,
+      "rewards/reward_len": -192.5875,
+      "step": 380
+    },
+    {
+      "completion_length": 62.6375,
+      "epoch": 0.0033412724250783915,
+      "grad_norm": 1.9541772603988647,
+      "kl": 0.03603251725435257,
+      "learning_rate": 9.988890983133714e-07,
+      "loss": 0.0014,
+      "reward": -95.625,
+      "reward_std": 76.96504859924316,
+      "rewards/reward_len": -95.625,
+      "step": 390
+    },
+    {
+      "completion_length": 76.5625,
+      "epoch": 0.0034269460770034784,
+      "grad_norm": 2.578866720199585,
+      "kl": 0.036808890849351884,
+      "learning_rate": 9.988605404293962e-07,
+      "loss": 0.0015,
+      "reward": -159.45,
+      "reward_std": 81.1762622833252,
+      "rewards/reward_len": -159.45,
+      "step": 400
+    },
+    {
+      "completion_length": 72.0875,
+      "epoch": 0.0035126197289285653,
+      "grad_norm": 2.2174038887023926,
+      "kl": 0.03971981406211853,
+      "learning_rate": 9.988319825454213e-07,
+      "loss": 0.0016,
+      "reward": -125.9625,
+      "reward_std": 70.27665119171142,
+      "rewards/reward_len": -125.9625,
+      "step": 410
+    },
+    {
+      "completion_length": 70.15,
+      "epoch": 0.0035982933808536522,
+      "grad_norm": 2.2227845191955566,
+      "kl": 0.03956588953733444,
+      "learning_rate": 9.988034246614463e-07,
+      "loss": 0.0016,
+      "reward": -110.5125,
+      "reward_std": 68.30635490417481,
+      "rewards/reward_len": -110.5125,
+      "step": 420
+    },
+    {
+      "completion_length": 66.8,
+      "epoch": 0.003683967032778739,
+      "grad_norm": 2.6529383659362793,
+      "kl": 0.052050336450338366,
+      "learning_rate": 9.987748667774712e-07,
+      "loss": 0.0021,
+      "reward": -110.275,
+      "reward_std": 86.19376144409179,
+      "rewards/reward_len": -110.275,
+      "step": 430
+    },
+    {
+      "completion_length": 71.2375,
+      "epoch": 0.003769640684703826,
+      "grad_norm": 2.401475429534912,
+      "kl": 0.05769926495850086,
+      "learning_rate": 9.987463088934962e-07,
+      "loss": 0.0023,
+      "reward": -128.325,
+      "reward_std": 84.20119457244873,
+      "rewards/reward_len": -128.325,
+      "step": 440
+    },
+    {
+      "completion_length": 90.3875,
+      "epoch": 0.003855314336628913,
+      "grad_norm": 2.8859000205993652,
+      "kl": 0.043742291163653135,
+      "learning_rate": 9.98717751009521e-07,
+      "loss": 0.0017,
+      "reward": -181.6,
+      "reward_std": 75.12719230651855,
+      "rewards/reward_len": -181.6,
+      "step": 450
+    },
+    {
+      "completion_length": 67.7625,
+      "epoch": 0.003940987988554,
+      "grad_norm": 2.0204262733459473,
+      "kl": 0.0622777983546257,
+      "learning_rate": 9.98689193125546e-07,
+      "loss": 0.0025,
+      "reward": -107.1125,
+      "reward_std": 67.32327232360839,
+      "rewards/reward_len": -107.1125,
+      "step": 460
+    },
+    {
+      "completion_length": 66.6,
+      "epoch": 0.004026661640479087,
+      "grad_norm": 2.5924794673919678,
+      "kl": 0.056221196055412294,
+      "learning_rate": 9.986606352415711e-07,
+      "loss": 0.0022,
+      "reward": -114.975,
+      "reward_std": 75.61366195678711,
+      "rewards/reward_len": -114.975,
+      "step": 470
+    },
+    {
+      "completion_length": 63.25,
+      "epoch": 0.004112335292404174,
+      "grad_norm": 2.5766448974609375,
+      "kl": 0.06269333846867084,
+      "learning_rate": 9.98632077357596e-07,
+      "loss": 0.0025,
+      "reward": -87.025,
+      "reward_std": 75.8617115020752,
+      "rewards/reward_len": -87.025,
+      "step": 480
+    },
+    {
+      "completion_length": 54.6625,
+      "epoch": 0.004198008944329261,
+      "grad_norm": 3.059234142303467,
+      "kl": 0.06824085712432862,
+      "learning_rate": 9.98603519473621e-07,
+      "loss": 0.0027,
+      "reward": -58.3375,
+      "reward_std": 65.2177251815796,
+      "rewards/reward_len": -58.3375,
+      "step": 490
+    },
+    {
+      "completion_length": 78.8125,
+      "epoch": 0.004283682596254348,
+      "grad_norm": 2.9885940551757812,
+      "kl": 0.07582498777192086,
+      "learning_rate": 9.98574961589646e-07,
+      "loss": 0.003,
+      "reward": -143.9,
+      "reward_std": 63.89678440093994,
+      "rewards/reward_len": -143.9,
+      "step": 500
+    },
+    {
+      "completion_length": 58.8375,
+      "epoch": 0.004369356248179435,
+      "grad_norm": 2.7045488357543945,
+      "kl": 0.10633525140583515,
+      "learning_rate": 9.98546403705671e-07,
+      "loss": 0.0043,
+      "reward": -71.625,
+      "reward_std": 78.84037780761719,
+      "rewards/reward_len": -71.625,
+      "step": 510
+    },
+    {
+      "completion_length": 66.35,
+      "epoch": 0.004455029900104522,
+      "grad_norm": 2.9650001525878906,
+      "kl": 0.08462142385542393,
+      "learning_rate": 9.98517845821696e-07,
+      "loss": 0.0034,
+      "reward": -93.275,
+      "reward_std": 69.99792709350587,
+      "rewards/reward_len": -93.275,
+      "step": 520
+    },
+    {
+      "completion_length": 81.0125,
+      "epoch": 0.004540703552029609,
+      "grad_norm": 3.4145212173461914,
+      "kl": 0.08472472308203578,
+      "learning_rate": 9.98489287937721e-07,
+      "loss": 0.0034,
+      "reward": -150.125,
+      "reward_std": 64.46914005279541,
+      "rewards/reward_len": -150.125,
+      "step": 530
+    },
+    {
+      "completion_length": 58.9625,
+      "epoch": 0.004626377203954696,
+      "grad_norm": 2.866358757019043,
+      "kl": 0.11239302009344102,
+      "learning_rate": 9.984607300537458e-07,
+      "loss": 0.0045,
+      "reward": -73.075,
+      "reward_std": 66.32526473999023,
+      "rewards/reward_len": -73.075,
+      "step": 540
+    },
+    {
+      "completion_length": 65.3875,
+      "epoch": 0.004712050855879783,
+      "grad_norm": 3.9847896099090576,
+      "kl": 0.1188249222934246,
+      "learning_rate": 9.984321721697709e-07,
+      "loss": 0.0048,
+      "reward": -104.2125,
+      "reward_std": 84.05291709899902,
+      "rewards/reward_len": -104.2125,
+      "step": 550
+    },
+    {
+      "completion_length": 55.425,
+      "epoch": 0.00479772450780487,
+      "grad_norm": 1.9827405214309692,
+      "kl": 0.13096430897712708,
+      "learning_rate": 9.984036142857957e-07,
+      "loss": 0.0052,
+      "reward": -52.8875,
+      "reward_std": 76.0460901260376,
+      "rewards/reward_len": -52.8875,
+      "step": 560
+    },
+    {
+      "completion_length": 63.4,
+      "epoch": 0.004883398159729956,
+      "grad_norm": 2.7263333797454834,
+      "kl": 0.11985208839178085,
+      "learning_rate": 9.983750564018208e-07,
+      "loss": 0.0048,
+      "reward": -85.3,
+      "reward_std": 75.41236038208008,
+      "rewards/reward_len": -85.3,
+      "step": 570
+    },
+    {
+      "completion_length": 60.8125,
+      "epoch": 0.004969071811655044,
+      "grad_norm": 2.395049810409546,
+      "kl": 0.11624505072832107,
+      "learning_rate": 9.983464985178458e-07,
+      "loss": 0.0046,
+      "reward": -79.4,
+      "reward_std": 51.84059715270996,
+      "rewards/reward_len": -79.4,
+      "step": 580
+    },
+    {
+      "completion_length": 51.7125,
+      "epoch": 0.00505474546358013,
+      "grad_norm": 3.295987606048584,
+      "kl": 0.16594674922525882,
+      "learning_rate": 9.983179406338709e-07,
+      "loss": 0.0066,
+      "reward": -43.55,
+      "reward_std": 40.574602890014646,
+      "rewards/reward_len": -43.55,
+      "step": 590
+    },
+    {
+      "completion_length": 67.3375,
+      "epoch": 0.005140419115505218,
+      "grad_norm": 4.078490257263184,
+      "kl": 0.13739465102553367,
+      "learning_rate": 9.982893827498957e-07,
+      "loss": 0.0055,
+      "reward": -98.5625,
+      "reward_std": 72.55281829833984,
+      "rewards/reward_len": -98.5625,
+      "step": 600
+    },
+    {
+      "completion_length": 46.9625,
+      "epoch": 0.005226092767430304,
+      "grad_norm": 4.195443153381348,
+      "kl": 0.16595774739980698,
+      "learning_rate": 9.982608248659207e-07,
+      "loss": 0.0066,
+      "reward": -25.4375,
+      "reward_std": 67.01236572265626,
+      "rewards/reward_len": -25.4375,
+      "step": 610
+    },
+    {
+      "completion_length": 48.8625,
+      "epoch": 0.0053117664193553914,
+      "grad_norm": 2.9587647914886475,
+      "kl": 0.19849314391613007,
+      "learning_rate": 9.982322669819456e-07,
+      "loss": 0.0079,
+      "reward": -32.075,
+      "reward_std": 57.74148025512695,
+      "rewards/reward_len": -32.075,
+      "step": 620
+    },
+    {
+      "completion_length": 55.6125,
+      "epoch": 0.005397440071280479,
+      "grad_norm": 3.3368797302246094,
+      "kl": 0.1759088508784771,
+      "learning_rate": 9.982037090979706e-07,
+      "loss": 0.007,
+      "reward": -60.65,
+      "reward_std": 58.34819984436035,
+      "rewards/reward_len": -60.65,
+      "step": 630
+    },
+    {
+      "completion_length": 51.225,
+      "epoch": 0.005483113723205565,
+      "grad_norm": 2.436890125274658,
+      "kl": 0.19748679026961327,
+      "learning_rate": 9.981751512139957e-07,
+      "loss": 0.0079,
+      "reward": -42.5,
+      "reward_std": 70.55487613677978,
+      "rewards/reward_len": -42.5,
+      "step": 640
+    },
+    {
+      "completion_length": 49.6875,
+      "epoch": 0.005568787375130653,
+      "grad_norm": 2.7104854583740234,
+      "kl": 0.2179098792374134,
+      "learning_rate": 9.981465933300205e-07,
+      "loss": 0.0087,
+      "reward": -41.025,
+      "reward_std": 67.48074893951416,
+      "rewards/reward_len": -41.025,
+      "step": 650
+    },
+    {
+      "completion_length": 45.3625,
+      "epoch": 0.005654461027055739,
+      "grad_norm": 2.9114573001861572,
+      "kl": 0.292061947286129,
+      "learning_rate": 9.981180354460456e-07,
+      "loss": 0.0117,
+      "reward": -22.2875,
+      "reward_std": 67.80285606384277,
+      "rewards/reward_len": -22.2875,
+      "step": 660
+    },
+    {
+      "completion_length": 43.25,
+      "epoch": 0.0057401346789808265,
+      "grad_norm": 4.684057712554932,
+      "kl": 0.23551913797855378,
+      "learning_rate": 9.980894775620706e-07,
+      "loss": 0.0094,
+      "reward": -7.725,
+      "reward_std": 54.08637924194336,
+      "rewards/reward_len": -7.725,
+      "step": 670
+    },
+    {
+      "completion_length": 39.75,
+      "epoch": 0.005825808330905913,
+      "grad_norm": 3.1786115169525146,
+      "kl": 0.32834408432245255,
+      "learning_rate": 9.980609196780954e-07,
+      "loss": 0.0131,
+      "reward": 6.0375,
+      "reward_std": 58.28718643188476,
+      "rewards/reward_len": 6.0375,
+      "step": 680
+    },
+    {
+      "completion_length": 49.025,
+      "epoch": 0.005911481982831,
+      "grad_norm": 4.3211870193481445,
+      "kl": 0.30903512686491014,
+      "learning_rate": 9.980323617941205e-07,
+      "loss": 0.0124,
+      "reward": -36.75,
+      "reward_std": 74.52720184326172,
+      "rewards/reward_len": -36.75,
+      "step": 690
+    },
+    {
+      "completion_length": 37.65,
+      "epoch": 0.005997155634756087,
+      "grad_norm": 2.111032247543335,
+      "kl": 0.32604690343141557,
+      "learning_rate": 9.980038039101453e-07,
+      "loss": 0.013,
+      "reward": 18.4125,
+      "reward_std": 50.093895721435544,
+      "rewards/reward_len": 18.4125,
+      "step": 700
+    },
+    {
+      "completion_length": 33.475,
+      "epoch": 0.006082829286681174,
+      "grad_norm": 4.225799560546875,
+      "kl": 0.3358478359878063,
+      "learning_rate": 9.979752460261704e-07,
+      "loss": 0.0134,
+      "reward": 29.3375,
+      "reward_std": 45.97467594146728,
+      "rewards/reward_len": 29.3375,
+      "step": 710
+    },
+    {
+      "completion_length": 36.1875,
+      "epoch": 0.006168502938606261,
+      "grad_norm": 3.743486166000366,
+      "kl": 0.3429659426212311,
+      "learning_rate": 9.979466881421954e-07,
+      "loss": 0.0137,
+      "reward": 14.65,
+      "reward_std": 42.70287170410156,
+      "rewards/reward_len": 14.65,
+      "step": 720
+    },
+    {
+      "completion_length": 36.1375,
+      "epoch": 0.006254176590531348,
+      "grad_norm": 3.634866952896118,
+      "kl": 0.41900385320186617,
+      "learning_rate": 9.979181302582205e-07,
+      "loss": 0.0168,
+      "reward": 21.025,
+      "reward_std": 38.98911609649658,
+      "rewards/reward_len": 21.025,
+      "step": 730
+    },
+    {
+      "completion_length": 38.7,
+      "epoch": 0.006339850242456435,
+      "grad_norm": 5.735452175140381,
+      "kl": 0.3897597849369049,
+      "learning_rate": 9.978895723742453e-07,
+      "loss": 0.0156,
+      "reward": 11.05,
+      "reward_std": 49.65783863067627,
+      "rewards/reward_len": 11.05,
+      "step": 740
+    },
+    {
+      "completion_length": 38.1125,
+      "epoch": 0.006425523894381522,
+      "grad_norm": 2.927375555038452,
+      "kl": 0.435553178191185,
+      "learning_rate": 9.978610144902703e-07,
+      "loss": 0.0174,
+      "reward": 8.8125,
+      "reward_std": 55.300292015075684,
+      "rewards/reward_len": 8.8125,
+      "step": 750
+    },
+    {
+      "completion_length": 29.675,
+      "epoch": 0.006511197546306609,
+      "grad_norm": 3.246641159057617,
+      "kl": 0.43810410499572755,
+      "learning_rate": 9.978324566062952e-07,
+      "loss": 0.0175,
+      "reward": 40.5375,
+      "reward_std": 35.55714553594589,
+      "rewards/reward_len": 40.5375,
+      "step": 760
+    },
+    {
+      "completion_length": 32.3125,
+      "epoch": 0.006596871198231696,
+      "grad_norm": 3.6894195079803467,
+      "kl": 0.5961927145719528,
+      "learning_rate": 9.978038987223202e-07,
+      "loss": 0.0238,
+      "reward": 28.3625,
+      "reward_std": 40.15065488815308,
+      "rewards/reward_len": 28.3625,
+      "step": 770
+    },
+    {
+      "completion_length": 33.0625,
+      "epoch": 0.006682544850156783,
+      "grad_norm": 4.511296272277832,
+      "kl": 0.5751866042613983,
+      "learning_rate": 9.977753408383453e-07,
+      "loss": 0.023,
+      "reward": 27.575,
+      "reward_std": 46.356824684143064,
+      "rewards/reward_len": 27.575,
+      "step": 780
+    },
+    {
+      "completion_length": 33.0375,
+      "epoch": 0.0067682185020818695,
+      "grad_norm": 3.548011302947998,
+      "kl": 0.42039217352867125,
+      "learning_rate": 9.9774678295437e-07,
+      "loss": 0.0168,
+      "reward": 26.5125,
+      "reward_std": 46.602904319763184,
+      "rewards/reward_len": 26.5125,
+      "step": 790
+    },
+    {
+      "completion_length": 26.85,
+      "epoch": 0.006853892154006957,
+      "grad_norm": 5.255430221557617,
+      "kl": 0.5921719163656235,
+      "learning_rate": 9.977182250703951e-07,
+      "loss": 0.0237,
+      "reward": 60.1375,
+      "reward_std": 30.120204424858095,
+      "rewards/reward_len": 60.1375,
+      "step": 800
+    },
+    {
+      "completion_length": 31.825,
+      "epoch": 0.006939565805932043,
+      "grad_norm": 2.8932251930236816,
+      "kl": 0.5926052063703537,
+      "learning_rate": 9.976896671864202e-07,
+      "loss": 0.0237,
+      "reward": 37.225,
+      "reward_std": 37.70037169456482,
+      "rewards/reward_len": 37.225,
+      "step": 810
+    },
+    {
+      "completion_length": 43.175,
+      "epoch": 0.007025239457857131,
+      "grad_norm": 2.886444091796875,
+      "kl": 0.5511217802762985,
+      "learning_rate": 9.97661109302445e-07,
+      "loss": 0.022,
+      "reward": -2.45,
+      "reward_std": 56.478550338745116,
+      "rewards/reward_len": -2.45,
+      "step": 820
+    },
+    {
+      "completion_length": 30.2625,
+      "epoch": 0.007110913109782218,
+      "grad_norm": 2.814298629760742,
+      "kl": 0.7488752245903015,
+      "learning_rate": 9.9763255141847e-07,
+      "loss": 0.03,
+      "reward": 43.575,
+      "reward_std": 47.09467077255249,
+      "rewards/reward_len": 43.575,
+      "step": 830
+    },
+    {
+      "completion_length": 25.85,
+      "epoch": 0.0071965867617073045,
+      "grad_norm": 5.516399383544922,
+      "kl": 0.7932253420352936,
+      "learning_rate": 9.97603993534495e-07,
+      "loss": 0.0317,
+      "reward": 63.85,
+      "reward_std": 25.45775556564331,
+      "rewards/reward_len": 63.85,
+      "step": 840
+    },
+    {
+      "completion_length": 27.95,
+      "epoch": 0.007282260413632392,
+      "grad_norm": 4.706915378570557,
+      "kl": 0.6608895510435104,
+      "learning_rate": 9.9757543565052e-07,
+      "loss": 0.0264,
+      "reward": 52.7,
+      "reward_std": 39.781547927856444,
+      "rewards/reward_len": 52.7,
+      "step": 850
+    },
+    {
+      "completion_length": 26.9125,
+      "epoch": 0.007367934065557478,
+      "grad_norm": 14.130315780639648,
+      "kl": 0.6030549615621567,
+      "learning_rate": 9.97546877766545e-07,
+      "loss": 0.0241,
+      "reward": 62.875,
+      "reward_std": 28.904275512695314,
+      "rewards/reward_len": 62.875,
+      "step": 860
+    },
+    {
+      "completion_length": 27.35,
+      "epoch": 0.007453607717482566,
+      "grad_norm": 5.187411785125732,
+      "kl": 0.7092643141746521,
+      "learning_rate": 9.9751831988257e-07,
+      "loss": 0.0284,
+      "reward": 47.6375,
+      "reward_std": 32.484121799468994,
+      "rewards/reward_len": 47.6375,
+      "step": 870
+    },
+    {
+      "completion_length": 46.7875,
+      "epoch": 0.007539281369407652,
+      "grad_norm": 2.9746391773223877,
+      "kl": 0.6793952610343694,
+      "learning_rate": 9.974897619985949e-07,
+      "loss": 0.0272,
+      "reward": -19.2625,
+      "reward_std": 51.298577308654785,
+      "rewards/reward_len": -19.2625,
+      "step": 880
+    },
+    {
+      "completion_length": 26.6375,
+      "epoch": 0.0076249550213327395,
+      "grad_norm": 4.254567623138428,
+      "kl": 0.7589318662881851,
+      "learning_rate": 9.9746120411462e-07,
+      "loss": 0.0304,
+      "reward": 57.5625,
+      "reward_std": 39.20279521942139,
+      "rewards/reward_len": 57.5625,
+      "step": 890
+    },
+    {
+      "completion_length": 24.1375,
+      "epoch": 0.007710628673257826,
+      "grad_norm": 4.250896453857422,
+      "kl": 0.8811014115810394,
+      "learning_rate": 9.974326462306448e-07,
+      "loss": 0.0352,
+      "reward": 73.8625,
+      "reward_std": 19.012751007080077,
+      "rewards/reward_len": 73.8625,
+      "step": 900
+    },
+    {
+      "completion_length": 25.4,
+      "epoch": 0.007796302325182913,
+      "grad_norm": 3.278289556503296,
+      "kl": 0.8711494505405426,
+      "learning_rate": 9.974040883466698e-07,
+      "loss": 0.0348,
+      "reward": 69.4375,
+      "reward_std": 24.089942455291748,
+      "rewards/reward_len": 69.4375,
+      "step": 910
+    },
+    {
+      "completion_length": 25.275,
+      "epoch": 0.007881975977108,
+      "grad_norm": 4.476153373718262,
+      "kl": 0.8596636682748795,
+      "learning_rate": 9.973755304626946e-07,
+      "loss": 0.0344,
+      "reward": 63.3875,
+      "reward_std": 20.410217094421387,
+      "rewards/reward_len": 63.3875,
+      "step": 920
+    },
+    {
+      "completion_length": 24.8625,
+      "epoch": 0.007967649629033088,
+      "grad_norm": 5.825425624847412,
+      "kl": 1.0523226261138916,
+      "learning_rate": 9.9734697257872e-07,
+      "loss": 0.0421,
+      "reward": 66.95,
+      "reward_std": 25.993260192871094,
+      "rewards/reward_len": 66.95,
+      "step": 930
+    },
+    {
+      "completion_length": 23.0875,
+      "epoch": 0.008053323280958175,
+      "grad_norm": 4.372145652770996,
+      "kl": 0.9800844460725784,
+      "learning_rate": 9.973184146947447e-07,
+      "loss": 0.0392,
+      "reward": 67.175,
+      "reward_std": 27.855006313323976,
+      "rewards/reward_len": 67.175,
+      "step": 940
+    },
+    {
+      "completion_length": 23.325,
+      "epoch": 0.008138996932883261,
+      "grad_norm": 5.127079010009766,
+      "kl": 0.9943849921226502,
+      "learning_rate": 9.972898568107698e-07,
+      "loss": 0.0398,
+      "reward": 78.1125,
+      "reward_std": 17.191841506958006,
+      "rewards/reward_len": 78.1125,
+      "step": 950
+    },
+    {
+      "completion_length": 22.9375,
+      "epoch": 0.008224670584808347,
+      "grad_norm": 4.183666229248047,
+      "kl": 1.068257564306259,
+      "learning_rate": 9.972612989267946e-07,
+      "loss": 0.0427,
+      "reward": 79.1625,
+      "reward_std": 18.828048992156983,
+      "rewards/reward_len": 79.1625,
+      "step": 960
+    },
+    {
+      "completion_length": 22.9375,
+      "epoch": 0.008310344236733436,
+      "grad_norm": 4.63860559463501,
+      "kl": 1.0732409298419952,
+      "learning_rate": 9.972327410428197e-07,
+      "loss": 0.0429,
+      "reward": 71.4875,
+      "reward_std": 22.58247709274292,
+      "rewards/reward_len": 71.4875,
+      "step": 970
+    },
+    {
+      "completion_length": 25.125,
+      "epoch": 0.008396017888658522,
+      "grad_norm": 5.824211597442627,
+      "kl": 0.9688010692596436,
+      "learning_rate": 9.972041831588445e-07,
+      "loss": 0.0388,
+      "reward": 67.6,
+      "reward_std": 23.308015012741087,
+      "rewards/reward_len": 67.6,
+      "step": 980
+    },
+    {
+      "completion_length": 22.6375,
+      "epoch": 0.008481691540583609,
+      "grad_norm": 2.8945517539978027,
+      "kl": 1.118459904193878,
+      "learning_rate": 9.971756252748695e-07,
+      "loss": 0.0447,
+      "reward": 81.375,
+      "reward_std": 17.22281708717346,
+      "rewards/reward_len": 81.375,
+      "step": 990
+    },
+    {
+      "completion_length": 21.275,
+      "epoch": 0.008567365192508695,
+      "grad_norm": 4.304519176483154,
+      "kl": 0.9363485395908355,
+      "learning_rate": 9.971470673908946e-07,
+      "loss": 0.0375,
+      "reward": 75.775,
+      "reward_std": 23.021865463256837,
+      "rewards/reward_len": 75.775,
+      "step": 1000
+    },
+    {
+      "completion_length": 18.8625,
+      "epoch": 0.008653038844433783,
+      "grad_norm": 4.940158843994141,
+      "kl": 1.0820651412010194,
+      "learning_rate": 9.971185095069196e-07,
+      "loss": 0.0433,
+      "reward": 76.725,
+      "reward_std": 19.4085599899292,
+      "rewards/reward_len": 76.725,
+      "step": 1010
+    },
+    {
+      "completion_length": 22.45,
+      "epoch": 0.00873871249635887,
+      "grad_norm": 6.257218837738037,
+      "kl": 1.3543362498283387,
+      "learning_rate": 9.970899516229445e-07,
+      "loss": 0.0542,
+      "reward": 73.275,
+      "reward_std": 21.16908242702484,
+      "rewards/reward_len": 73.275,
+      "step": 1020
+    },
+    {
+      "completion_length": 21.1875,
+      "epoch": 0.008824386148283956,
+      "grad_norm": 7.533606052398682,
+      "kl": 0.9111672580242157,
+      "learning_rate": 9.970613937389695e-07,
+      "loss": 0.0364,
+      "reward": 79.5375,
+      "reward_std": 17.676442378759383,
+      "rewards/reward_len": 79.5375,
+      "step": 1030
+    },
+    {
+      "completion_length": 17.7,
+      "epoch": 0.008910059800209045,
+      "grad_norm": 6.559030532836914,
+      "kl": 1.283721137046814,
+      "learning_rate": 9.970328358549944e-07,
+      "loss": 0.0513,
+      "reward": 86.375,
+      "reward_std": 15.953600931167603,
+      "rewards/reward_len": 86.375,
+      "step": 1040
+    },
+    {
+      "completion_length": 21.5125,
+      "epoch": 0.008995733452134131,
+      "grad_norm": 9.517870903015137,
+      "kl": 1.2262637436389923,
+      "learning_rate": 9.970042779710194e-07,
+      "loss": 0.0491,
+      "reward": 75.275,
+      "reward_std": 21.458053398132325,
+      "rewards/reward_len": 75.275,
+      "step": 1050
+    },
+    {
+      "completion_length": 21.75,
+      "epoch": 0.009081407104059218,
+      "grad_norm": 12.434151649475098,
+      "kl": 1.324447250366211,
+      "learning_rate": 9.969757200870442e-07,
+      "loss": 0.053,
+      "reward": 81.1125,
+      "reward_std": 16.958592414855957,
+      "rewards/reward_len": 81.1125,
+      "step": 1060
+    },
+    {
+      "completion_length": 22.225,
+      "epoch": 0.009167080755984304,
+      "grad_norm": 5.839966297149658,
+      "kl": 1.3111671924591064,
+      "learning_rate": 9.969471622030695e-07,
+      "loss": 0.0524,
+      "reward": 74.425,
+      "reward_std": 27.983480072021486,
+      "rewards/reward_len": 74.425,
+      "step": 1070
+    },
+    {
+      "completion_length": 20.3,
+      "epoch": 0.009252754407909392,
+      "grad_norm": 5.334268093109131,
+      "kl": 1.2456512570381164,
+      "learning_rate": 9.969186043190943e-07,
+      "loss": 0.0498,
+      "reward": 81.6,
+      "reward_std": 18.58235924243927,
+      "rewards/reward_len": 81.6,
+      "step": 1080
+    },
+    {
+      "completion_length": 22.3125,
+      "epoch": 0.009338428059834479,
+      "grad_norm": 4.366571426391602,
+      "kl": 1.3097686409950255,
+      "learning_rate": 9.968900464351194e-07,
+      "loss": 0.0524,
+      "reward": 77.85,
+      "reward_std": 20.753051567077637,
+      "rewards/reward_len": 77.85,
+      "step": 1090
+    },
+    {
+      "completion_length": 17.3875,
+      "epoch": 0.009424101711759565,
+      "grad_norm": 6.299209117889404,
+      "kl": 1.508902084827423,
+      "learning_rate": 9.968614885511442e-07,
+      "loss": 0.0604,
+      "reward": 83.225,
+      "reward_std": 13.49931423664093,
+      "rewards/reward_len": 83.225,
+      "step": 1100
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 350166,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 50,
+  "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": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:634da9b317c6946bc1afcc076ed24dd9d7e174e0648a431f51ba355afa1762ff
+size 5752