Upload 9 files

Browse files

Files changed (8) hide show

added_tokens.json +24 -28
config.json +29 -29
generation_config.json +14 -14
special_tokens_map.json +31 -31
tokenizer.json +2 -2
tokenizer_config.json +209 -241
trainer_state.json +1278 -1278
vocab.json +0 -0

added_tokens.json CHANGED Viewed

@@ -1,28 +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,
-    "</think>": 151668,
-    "</tool_response>": 151666,
-    "<think>": 151667,
-    "<tool_response>": 151665
-}

+{
+  "</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 CHANGED Viewed

@@ -1,29 +1,29 @@
-{
-    "_name_or_path": "NovaSky-AI/Sky-T1-32B-Preview",
-    "architectures": [
-        "Qwen2ForCausalLM"
-    ],
-    "attention_dropout": 0.0,
-    "bos_token_id": 151643,
-    "eos_token_id": 151645,
-    "hidden_act": "silu",
-    "hidden_size": 5120,
-    "initializer_range": 0.02,
-    "intermediate_size": 27648,
-    "max_position_embeddings": 40960,
-    "max_window_layers": 64,
-    "model_type": "qwen2",
-    "num_attention_heads": 40,
-    "num_hidden_layers": 64,
-    "num_key_value_heads": 8,
-    "rms_norm_eps": 1e-05,
-    "rope_scaling": null,
-    "rope_theta": 1000000.0,
-    "sliding_window": 32768,
-    "tie_word_embeddings": false,
-    "torch_dtype": "bfloat16",
-    "transformers_version": "4.43.1",
-    "use_cache": true,
-    "use_sliding_window": false,
-    "vocab_size": 151665
-}

+{
+  "_name_or_path": "NovaSky-AI/Sky-T1-32B-Preview",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 27648,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 70,
+  "model_type": "qwen2",
+  "num_attention_heads": 40,
+  "num_hidden_layers": 64,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.1",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

generation_config.json CHANGED Viewed

@@ -1,14 +1,14 @@
-{
-    "bos_token_id": 151643,
-    "do_sample": true,
-    "eos_token_id": [
-        151645,
-        151643
-    ],
-    "pad_token_id": 151643,
-    "repetition_penalty": 1.0,
-    "temperature": 0.6,
-    "top_k": 40,
-    "top_p": 0.95,
-    "transformers_version": "4.45.2"
-}

+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.46.1"
+}

special_tokens_map.json CHANGED Viewed

@@ -1,31 +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
-    }
-}

+{
+  "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": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3cd6bd4dabe832e58465a15a6c1a6e07684b07b1fc4ae54330926771eff2671
-size 9154434

 version https://git-lfs.github.com/spec/v1
+oid sha256:6ae8594f633daa21dc0e04606279856d0ccb8abb64bb7a5a93e72531fb12b9d9
+size 11421890

tokenizer_config.json CHANGED Viewed

@@ -1,241 +1,209 @@
-{
-    "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    {%- else %}\n        {{- '' }}\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  {%- endif %}\n{%- endif %}\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\" and not message.tool_calls %}\n        {%- set content = message.content %}\n        {%- if not loop.last %}\n            {%- set content = message.content.split('</think>')[-1].lstrip('\\n    ') %}\n        {%- endif %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set content = message.content %}\n        {%- if not loop.last %}\n            {%- set content = message.content.split('</think>')[-1].lstrip('\\n') %}\n        {%- endif %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + 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<think>\\n' }}\n{%- endif %}\n",
-    "clean_up_tokenization_spaces": false,
-    "eos_token": "<|im_end|>",
-    "errors": "replace",
-    "extra_special_tokens": {},
-    "model_max_length": 131072,
-    "pad_token": "<|endoftext|>",
-    "padding_side": "right",
-    "split_special_tokens": false,
-    "tokenizer_class": "Qwen2Tokenizer",
-    "unk_token": null
-}

+{
+  "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": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json CHANGED Viewed

@@ -1,1278 +1,1278 @@
-{
-    "best_metric": null,
-    "best_model_checkpoint": null,
-    "epoch": 0.996,
-    "eval_steps": 500,
-    "global_step": 83,
-    "is_hyper_param_search": false,
-    "is_local_process_zero": true,
-    "is_world_process_zero": true,
-    "log_history": [
-        {
-            "epoch": 0.012,
-            "grad_norm": 0.8640882968902588,
-            "learning_rate": 5.555555555555555e-08,
-            "logits/chosen": -1.495650291442871,
-            "logits/rejected": -1.3535889387130737,
-            "logps/chosen": -0.10173828899860382,
-            "logps/rejected": -0.08792766183614731,
-            "loss": 0.8717,
-            "rewards/accuracies": 0.3333333730697632,
-            "rewards/chosen": -0.20347657799720764,
-            "rewards/margins": -0.027621246874332428,
-            "rewards/rejected": -0.17585532367229462,
-            "step": 1
-        },
-        {
-            "epoch": 0.024,
-            "grad_norm": 0.979682981967926,
-            "learning_rate": 1.111111111111111e-07,
-            "logits/chosen": -1.443913221359253,
-            "logits/rejected": -1.3641024827957153,
-            "logps/chosen": -0.10752908140420914,
-            "logps/rejected": -0.08240076899528503,
-            "loss": 0.8846,
-            "rewards/accuracies": 0.2187500149011612,
-            "rewards/chosen": -0.21505816280841827,
-            "rewards/margins": -0.0502566322684288,
-            "rewards/rejected": -0.16480153799057007,
-            "step": 2
-        },
-        {
-            "epoch": 0.036,
-            "grad_norm": 0.9309831857681274,
-            "learning_rate": 1.6666666666666665e-07,
-            "logits/chosen": -1.4732969999313354,
-            "logits/rejected": -1.376213550567627,
-            "logps/chosen": -0.10481980443000793,
-            "logps/rejected": -0.07933054119348526,
-            "loss": 0.8848,
-            "rewards/accuracies": 0.28125,
-            "rewards/chosen": -0.20963960886001587,
-            "rewards/margins": -0.050978537648916245,
-            "rewards/rejected": -0.15866108238697052,
-            "step": 3
-        },
-        {
-            "epoch": 0.048,
-            "grad_norm": 0.9682196974754333,
-            "learning_rate": 2.222222222222222e-07,
-            "logits/chosen": -1.5285106897354126,
-            "logits/rejected": -1.3950246572494507,
-            "logps/chosen": -0.11225953698158264,
-            "logps/rejected": -0.08694332838058472,
-            "loss": 0.8849,
-            "rewards/accuracies": 0.3229166865348816,
-            "rewards/chosen": -0.22451907396316528,
-            "rewards/margins": -0.05063238739967346,
-            "rewards/rejected": -0.17388665676116943,
-            "step": 4
-        },
-        {
-            "epoch": 0.06,
-            "grad_norm": 1.0820338726043701,
-            "learning_rate": 2.7777777777777776e-07,
-            "logits/chosen": -1.494052767753601,
-            "logits/rejected": -1.3356661796569824,
-            "logps/chosen": -0.12428013980388641,
-            "logps/rejected": -0.09185128659009933,
-            "loss": 0.8968,
-            "rewards/accuracies": 0.322916716337204,
-            "rewards/chosen": -0.24856027960777283,
-            "rewards/margins": -0.06485769897699356,
-            "rewards/rejected": -0.18370257318019867,
-            "step": 5
-        },
-        {
-            "epoch": 0.072,
-            "grad_norm": 1.0573477745056152,
-            "learning_rate": 3.333333333333333e-07,
-            "logits/chosen": -1.4825295209884644,
-            "logits/rejected": -1.3897716999053955,
-            "logps/chosen": -0.11380095779895782,
-            "logps/rejected": -0.08133920282125473,
-            "loss": 0.8938,
-            "rewards/accuracies": 0.3020833432674408,
-            "rewards/chosen": -0.22760191559791565,
-            "rewards/margins": -0.06492353230714798,
-            "rewards/rejected": -0.16267840564250946,
-            "step": 6
-        },
-        {
-            "epoch": 0.084,
-            "grad_norm": 0.9294664263725281,
-            "learning_rate": 3.888888888888889e-07,
-            "logits/chosen": -1.5021216869354248,
-            "logits/rejected": -1.3295238018035889,
-            "logps/chosen": -0.099081851541996,
-            "logps/rejected": -0.08119318634271622,
-            "loss": 0.8761,
-            "rewards/accuracies": 0.2916666567325592,
-            "rewards/chosen": -0.198163703083992,
-            "rewards/margins": -0.03577733412384987,
-            "rewards/rejected": -0.16238637268543243,
-            "step": 7
-        },
-        {
-            "epoch": 0.096,
-            "grad_norm": 1.0329580307006836,
-            "learning_rate": 4.444444444444444e-07,
-            "logits/chosen": -1.5048794746398926,
-            "logits/rejected": -1.377795934677124,
-            "logps/chosen": -0.12124787271022797,
-            "logps/rejected": -0.08464659005403519,
-            "loss": 0.9015,
-            "rewards/accuracies": 0.3125,
-            "rewards/chosen": -0.24249574542045593,
-            "rewards/margins": -0.07320256531238556,
-            "rewards/rejected": -0.16929318010807037,
-            "step": 8
-        },
-        {
-            "epoch": 0.108,
-            "grad_norm": 0.9767515063285828,
-            "learning_rate": 5e-07,
-            "logits/chosen": -1.5186768770217896,
-            "logits/rejected": -1.3877286911010742,
-            "logps/chosen": -0.10737170279026031,
-            "logps/rejected": -0.0823729932308197,
-            "loss": 0.8847,
-            "rewards/accuracies": 0.3020833432674408,
-            "rewards/chosen": -0.21474340558052063,
-            "rewards/margins": -0.04999742656946182,
-            "rewards/rejected": -0.1647459864616394,
-            "step": 9
-        },
-        {
-            "epoch": 0.12,
-            "grad_norm": 0.9653743505477905,
-            "learning_rate": 4.997747415511704e-07,
-            "logits/chosen": -1.4915320873260498,
-            "logits/rejected": -1.3450301885604858,
-            "logps/chosen": -0.10813666880130768,
-            "logps/rejected": -0.08507634699344635,
-            "loss": 0.8822,
-            "rewards/accuracies": 0.2291666716337204,
-            "rewards/chosen": -0.21627333760261536,
-            "rewards/margins": -0.04612065851688385,
-            "rewards/rejected": -0.1701526939868927,
-            "step": 10
-        },
-        {
-            "epoch": 0.132,
-            "grad_norm": 0.9533908367156982,
-            "learning_rate": 4.990993721356315e-07,
-            "logits/chosen": -1.4848368167877197,
-            "logits/rejected": -1.3539990186691284,
-            "logps/chosen": -0.11860796809196472,
-            "logps/rejected": -0.08763512969017029,
-            "loss": 0.8931,
-            "rewards/accuracies": 0.25,
-            "rewards/chosen": -0.23721593618392944,
-            "rewards/margins": -0.061945684254169464,
-            "rewards/rejected": -0.17527025938034058,
-            "step": 11
-        },
-        {
-            "epoch": 0.144,
-            "grad_norm": 0.9020726084709167,
-            "learning_rate": 4.979751088147191e-07,
-            "logits/chosen": -1.4897594451904297,
-            "logits/rejected": -1.3646252155303955,
-            "logps/chosen": -0.09957993775606155,
-            "logps/rejected": -0.08218874782323837,
-            "loss": 0.8759,
-            "rewards/accuracies": 0.3958333432674408,
-            "rewards/chosen": -0.1991598755121231,
-            "rewards/margins": -0.034782394766807556,
-            "rewards/rejected": -0.16437749564647675,
-            "step": 12
-        },
-        {
-            "epoch": 0.156,
-            "grad_norm": 1.0148029327392578,
-            "learning_rate": 4.964039775869271e-07,
-            "logits/chosen": -1.4809292554855347,
-            "logits/rejected": -1.3655236959457397,
-            "logps/chosen": -0.11366236209869385,
-            "logps/rejected": -0.0875682383775711,
-            "loss": 0.8868,
-            "rewards/accuracies": 0.3645833432674408,
-            "rewards/chosen": -0.2273247241973877,
-            "rewards/margins": -0.05218825116753578,
-            "rewards/rejected": -0.1751364767551422,
-            "step": 13
-        },
-        {
-            "epoch": 0.168,
-            "grad_norm": 1.334598422050476,
-            "learning_rate": 4.943888097369216e-07,
-            "logits/chosen": -1.4804601669311523,
-            "logits/rejected": -1.3518431186676025,
-            "logps/chosen": -0.13359910249710083,
-            "logps/rejected": -0.08674132823944092,
-            "loss": 0.9236,
-            "rewards/accuracies": 0.2395833432674408,
-            "rewards/chosen": -0.26719820499420166,
-            "rewards/margins": -0.09371551126241684,
-            "rewards/rejected": -0.17348265647888184,
-            "step": 14
-        },
-        {
-            "epoch": 0.18,
-            "grad_norm": 0.9834568500518799,
-            "learning_rate": 4.919332367333748e-07,
-            "logits/chosen": -1.4974645376205444,
-            "logits/rejected": -1.3400018215179443,
-            "logps/chosen": -0.10907550156116486,
-            "logps/rejected": -0.08122590184211731,
-            "loss": 0.8879,
-            "rewards/accuracies": 0.21875,
-            "rewards/chosen": -0.2181510031223297,
-            "rewards/margins": -0.05569921433925629,
-            "rewards/rejected": -0.16245180368423462,
-            "step": 15
-        },
-        {
-            "epoch": 0.192,
-            "grad_norm": 0.9361069798469543,
-            "learning_rate": 4.890416836848127e-07,
-            "logits/chosen": -1.4610369205474854,
-            "logits/rejected": -1.369497299194336,
-            "logps/chosen": -0.10335300862789154,
-            "logps/rejected": -0.08525583893060684,
-            "loss": 0.8765,
-            "rewards/accuracies": 0.3541666865348816,
-            "rewards/chosen": -0.20670601725578308,
-            "rewards/margins": -0.03619435429573059,
-            "rewards/rejected": -0.17051167786121368,
-            "step": 16
-        },
-        {
-            "epoch": 0.204,
-            "grad_norm": 0.910740077495575,
-            "learning_rate": 4.85719361365271e-07,
-            "logits/chosen": -1.5087649822235107,
-            "logits/rejected": -1.3652905225753784,
-            "logps/chosen": -0.09960392862558365,
-            "logps/rejected": -0.08507402241230011,
-            "loss": 0.8719,
-            "rewards/accuracies": 0.3541666865348816,
-            "rewards/chosen": -0.1992078572511673,
-            "rewards/margins": -0.029059793800115585,
-            "rewards/rejected": -0.17014804482460022,
-            "step": 17
-        },
-        {
-            "epoch": 0.216,
-            "grad_norm": 0.9233217239379883,
-            "learning_rate": 4.819722568241273e-07,
-            "logits/chosen": -1.4695565700531006,
-            "logits/rejected": -1.336183786392212,
-            "logps/chosen": -0.10343371331691742,
-            "logps/rejected": -0.0857667475938797,
-            "loss": 0.8769,
-            "rewards/accuracies": 0.4166666865348816,
-            "rewards/chosen": -0.20686742663383484,
-            "rewards/margins": -0.03533393144607544,
-            "rewards/rejected": -0.1715334951877594,
-            "step": 18
-        },
-        {
-            "epoch": 0.228,
-            "grad_norm": 0.8909419775009155,
-            "learning_rate": 4.778071225970339e-07,
-            "logits/chosen": -1.4759807586669922,
-            "logits/rejected": -1.3495370149612427,
-            "logps/chosen": -0.10235996544361115,
-            "logps/rejected": -0.0851350948214531,
-            "loss": 0.8757,
-            "rewards/accuracies": 0.3229166567325592,
-            "rewards/chosen": -0.2047199308872223,
-            "rewards/margins": -0.03444972261786461,
-            "rewards/rejected": -0.1702701896429062,
-            "step": 19
-        },
-        {
-            "epoch": 0.24,
-            "grad_norm": 0.9479613304138184,
-            "learning_rate": 4.732314645373921e-07,
-            "logits/chosen": -1.4673627614974976,
-            "logits/rejected": -1.3651877641677856,
-            "logps/chosen": -0.10552840679883957,
-            "logps/rejected": -0.07911674678325653,
-            "loss": 0.886,
-            "rewards/accuracies": 0.28125,
-            "rewards/chosen": -0.21105681359767914,
-            "rewards/margins": -0.05282333493232727,
-            "rewards/rejected": -0.15823349356651306,
-            "step": 20
-        },
-        {
-            "epoch": 0.252,
-            "grad_norm": 0.854266345500946,
-            "learning_rate": 4.68253528290297e-07,
-            "logits/chosen": -1.4877924919128418,
-            "logits/rejected": -1.324515461921692,
-            "logps/chosen": -0.1002732366323471,
-            "logps/rejected": -0.09079625457525253,
-            "loss": 0.8664,
-            "rewards/accuracies": 0.3645833432674408,
-            "rewards/chosen": -0.2005464732646942,
-            "rewards/margins": -0.018953965976834297,
-            "rewards/rejected": -0.18159250915050507,
-            "step": 21
-        },
-        {
-            "epoch": 0.264,
-            "grad_norm": 0.9627916812896729,
-            "learning_rate": 4.6288228443332776e-07,
-            "logits/chosen": -1.466447114944458,
-            "logits/rejected": -1.3485172986984253,
-            "logps/chosen": -0.10300128906965256,
-            "logps/rejected": -0.08484764397144318,
-            "loss": 0.8763,
-            "rewards/accuracies": 0.3333333730697632,
-            "rewards/chosen": -0.20600257813930511,
-            "rewards/margins": -0.03630730137228966,
-            "rewards/rejected": -0.16969528794288635,
-            "step": 22
-        },
-        {
-            "epoch": 0.276,
-            "grad_norm": 0.8689938187599182,
-            "learning_rate": 4.571274123109605e-07,
-            "logits/chosen": -1.4124252796173096,
-            "logits/rejected": -1.3001394271850586,
-            "logps/chosen": -0.10588695108890533,
-            "logps/rejected": -0.08438257873058319,
-            "loss": 0.8809,
-            "rewards/accuracies": 0.2395833283662796,
-            "rewards/chosen": -0.21177390217781067,
-            "rewards/margins": -0.043008752167224884,
-            "rewards/rejected": -0.16876515746116638,
-            "step": 23
-        },
-        {
-            "epoch": 0.288,
-            "grad_norm": 0.9801927208900452,
-            "learning_rate": 4.5099928259173514e-07,
-            "logits/chosen": -1.4522674083709717,
-            "logits/rejected": -1.3461543321609497,
-            "logps/chosen": -0.1134054884314537,
-            "logps/rejected": -0.08870639652013779,
-            "loss": 0.886,
-            "rewards/accuracies": 0.2499999850988388,
-            "rewards/chosen": -0.2268109768629074,
-            "rewards/margins": -0.049398161470890045,
-            "rewards/rejected": -0.17741279304027557,
-            "step": 24
-        },
-        {
-            "epoch": 0.3,
-            "grad_norm": 0.9185568690299988,
-            "learning_rate": 4.4450893857960984e-07,
-            "logits/chosen": -1.4601349830627441,
-            "logits/rejected": -1.273384928703308,
-            "logps/chosen": -0.10715562850236893,
-            "logps/rejected": -0.09089934825897217,
-            "loss": 0.8746,
-            "rewards/accuracies": 0.375,
-            "rewards/chosen": -0.21431125700473785,
-            "rewards/margins": -0.03251257538795471,
-            "rewards/rejected": -0.18179869651794434,
-            "step": 25
-        },
-        {
-            "epoch": 0.312,
-            "grad_norm": 0.8864983320236206,
-            "learning_rate": 4.3766807631318105e-07,
-            "logits/chosen": -1.438947081565857,
-            "logits/rejected": -1.3086212873458862,
-            "logps/chosen": -0.10850708931684494,
-            "logps/rejected": -0.0892128050327301,
-            "loss": 0.8781,
-            "rewards/accuracies": 0.3229166865348816,
-            "rewards/chosen": -0.21701417863368988,
-            "rewards/margins": -0.038588590919971466,
-            "rewards/rejected": -0.1784256100654602,
-            "step": 26
-        },
-        {
-            "epoch": 0.324,
-            "grad_norm": 0.9518214464187622,
-            "learning_rate": 4.3048902348863106e-07,
-            "logits/chosen": -1.4363845586776733,
-            "logits/rejected": -1.3129128217697144,
-            "logps/chosen": -0.10809854418039322,
-            "logps/rejected": -0.08249183744192123,
-            "loss": 0.885,
-            "rewards/accuracies": 0.2812500298023224,
-            "rewards/chosen": -0.21619708836078644,
-            "rewards/margins": -0.05121342092752457,
-            "rewards/rejected": -0.16498367488384247,
-            "step": 27
-        },
-        {
-            "epoch": 0.336,
-            "grad_norm": 0.9291889667510986,
-            "learning_rate": 4.2298471724438653e-07,
-            "logits/chosen": -1.4329116344451904,
-            "logits/rejected": -1.3157219886779785,
-            "logps/chosen": -0.10641689598560333,
-            "logps/rejected": -0.0836237370967865,
-            "loss": 0.8825,
-            "rewards/accuracies": 0.3541666567325592,
-            "rewards/chosen": -0.21283379197120667,
-            "rewards/margins": -0.04558631405234337,
-            "rewards/rejected": -0.167247474193573,
-            "step": 28
-        },
-        {
-            "epoch": 0.348,
-            "grad_norm": 0.8883063197135925,
-            "learning_rate": 4.151686808475203e-07,
-            "logits/chosen": -1.4277637004852295,
-            "logits/rejected": -1.2657394409179688,
-            "logps/chosen": -0.1044892817735672,
-            "logps/rejected": -0.08842873573303223,
-            "loss": 0.8741,
-            "rewards/accuracies": 0.34375,
-            "rewards/chosen": -0.2089785635471344,
-            "rewards/margins": -0.03212107717990875,
-            "rewards/rejected": -0.17685747146606445,
-            "step": 29
-        },
-        {
-            "epoch": 0.36,
-            "grad_norm": 0.9316148161888123,
-            "learning_rate": 4.070549993239106e-07,
-            "logits/chosen": -1.3881316184997559,
-            "logits/rejected": -1.2497737407684326,
-            "logps/chosen": -0.11017285287380219,
-            "logps/rejected": -0.0868031308054924,
-            "loss": 0.8842,
-            "rewards/accuracies": 0.3645833432674408,
-            "rewards/chosen": -0.22034570574760437,
-            "rewards/margins": -0.046739429235458374,
-            "rewards/rejected": -0.1736062616109848,
-            "step": 30
-        },
-        {
-            "epoch": 0.372,
-            "grad_norm": 0.8768561482429504,
-            "learning_rate": 3.9865829407607166e-07,
-            "logits/chosen": -1.4147083759307861,
-            "logits/rejected": -1.2981321811676025,
-            "logps/chosen": -0.10104362666606903,
-            "logps/rejected": -0.09066756814718246,
-            "loss": 0.8673,
-            "rewards/accuracies": 0.3437500298023224,
-            "rewards/chosen": -0.20208725333213806,
-            "rewards/margins": -0.020752109587192535,
-            "rewards/rejected": -0.18133513629436493,
-            "step": 31
-        },
-        {
-            "epoch": 0.384,
-            "grad_norm": 1.0077239274978638,
-            "learning_rate": 3.8999369653439883e-07,
-            "logits/chosen": -1.4155137538909912,
-            "logits/rejected": -1.2743993997573853,
-            "logps/chosen": -0.10902610421180725,
-            "logps/rejected": -0.0863211378455162,
-            "loss": 0.8827,
-            "rewards/accuracies": 0.3020833432674408,
-            "rewards/chosen": -0.2180522084236145,
-            "rewards/margins": -0.0454099103808403,
-            "rewards/rejected": -0.1726422756910324,
-            "step": 32
-        },
-        {
-            "epoch": 0.396,
-            "grad_norm": 0.9207706451416016,
-            "learning_rate": 3.810768208893079e-07,
-            "logits/chosen": -1.3758422136306763,
-            "logits/rejected": -1.2915358543395996,
-            "logps/chosen": -0.1034766435623169,
-            "logps/rejected": -0.07977995276451111,
-            "loss": 0.883,
-            "rewards/accuracies": 0.3020833432674408,
-            "rewards/chosen": -0.2069532871246338,
-            "rewards/margins": -0.047393374145030975,
-            "rewards/rejected": -0.15955990552902222,
-            "step": 33
-        },
-        {
-            "epoch": 0.408,
-            "grad_norm": 0.9844294190406799,
-            "learning_rate": 3.7192373595340865e-07,
-            "logits/chosen": -1.442295789718628,
-            "logits/rejected": -1.2863086462020874,
-            "logps/chosen": -0.10710459202528,
-            "logps/rejected": -0.08674141019582748,
-            "loss": 0.8796,
-            "rewards/accuracies": 0.3020833432674408,
-            "rewards/chosen": -0.21420918405056,
-            "rewards/margins": -0.04072638228535652,
-            "rewards/rejected": -0.17348282039165497,
-            "step": 34
-        },
-        {
-            "epoch": 0.42,
-            "grad_norm": 0.9682800769805908,
-            "learning_rate": 3.625509362044183e-07,
-            "logits/chosen": -1.3701705932617188,
-            "logits/rejected": -1.2656924724578857,
-            "logps/chosen": -0.11000403016805649,
-            "logps/rejected": -0.08980907499790192,
-            "loss": 0.8803,
-            "rewards/accuracies": 0.3854166567325592,
-            "rewards/chosen": -0.22000806033611298,
-            "rewards/margins": -0.04038992151618004,
-            "rewards/rejected": -0.17961814999580383,
-            "step": 35
-        },
-        {
-            "epoch": 0.432,
-            "grad_norm": 0.9004039168357849,
-            "learning_rate": 3.529753120609982e-07,
-            "logits/chosen": -1.4099225997924805,
-            "logits/rejected": -1.252682089805603,
-            "logps/chosen": -0.09707480669021606,
-            "logps/rejected": -0.08859608322381973,
-            "loss": 0.8653,
-            "rewards/accuracies": 0.4479166865348816,
-            "rewards/chosen": -0.19414961338043213,
-            "rewards/margins": -0.016957445070147514,
-            "rewards/rejected": -0.17719216644763947,
-            "step": 36
-        },
-        {
-            "epoch": 0.444,
-            "grad_norm": 0.9568714499473572,
-            "learning_rate": 3.4321411944507714e-07,
-            "logits/chosen": -1.3612836599349976,
-            "logits/rejected": -1.2257184982299805,
-            "logps/chosen": -0.10623904317617416,
-            "logps/rejected": -0.09100136905908585,
-            "loss": 0.8734,
-            "rewards/accuracies": 0.4166666567325592,
-            "rewards/chosen": -0.21247808635234833,
-            "rewards/margins": -0.030475351959466934,
-            "rewards/rejected": -0.1820027381181717,
-            "step": 37
-        },
-        {
-            "epoch": 0.456,
-            "grad_norm": 0.8658465147018433,
-            "learning_rate": 3.332849486855144e-07,
-            "logits/chosen": -1.381141185760498,
-            "logits/rejected": -1.2495661973953247,
-            "logps/chosen": -0.09047922492027283,
-            "logps/rejected": -0.07975338399410248,
-            "loss": 0.8676,
-            "rewards/accuracies": 0.3958333432674408,
-            "rewards/chosen": -0.18095844984054565,
-            "rewards/margins": -0.021451696753501892,
-            "rewards/rejected": -0.15950676798820496,
-            "step": 38
-        },
-        {
-            "epoch": 0.468,
-            "grad_norm": 0.9716626405715942,
-            "learning_rate": 3.2320569281913754e-07,
-            "logits/chosen": -1.3790628910064697,
-            "logits/rejected": -1.251636028289795,
-            "logps/chosen": -0.10486049205064774,
-            "logps/rejected": -0.08277000486850739,
-            "loss": 0.883,
-            "rewards/accuracies": 0.3437500298023224,
-            "rewards/chosen": -0.20972098410129547,
-            "rewards/margins": -0.0441809706389904,
-            "rewards/rejected": -0.16554000973701477,
-            "step": 39
-        },
-        {
-            "epoch": 0.48,
-            "grad_norm": 0.948078989982605,
-            "learning_rate": 3.129945153462813e-07,
-            "logits/chosen": -1.3937729597091675,
-            "logits/rejected": -1.2695637941360474,
-            "logps/chosen": -0.09631801396608353,
-            "logps/rejected": -0.08585619181394577,
-            "loss": 0.8673,
-            "rewards/accuracies": 0.3854166865348816,
-            "rewards/chosen": -0.19263602793216705,
-            "rewards/margins": -0.020923633128404617,
-            "rewards/rejected": -0.17171238362789154,
-            "step": 40
-        },
-        {
-            "epoch": 0.492,
-            "grad_norm": 0.9445565342903137,
-            "learning_rate": 3.0266981749893154e-07,
-            "logits/chosen": -1.3988057374954224,
-            "logits/rejected": -1.2703980207443237,
-            "logps/chosen": -0.09746824949979782,
-            "logps/rejected": -0.0910995602607727,
-            "loss": 0.8624,
-            "rewards/accuracies": 0.40625,
-            "rewards/chosen": -0.19493649899959564,
-            "rewards/margins": -0.012737366370856762,
-            "rewards/rejected": -0.1821991205215454,
-            "step": 41
-        },
-        {
-            "epoch": 0.504,
-            "grad_norm": 1.080769419670105,
-            "learning_rate": 2.922502050804623e-07,
-            "logits/chosen": -1.4094312191009521,
-            "logits/rejected": -1.247127890586853,
-            "logps/chosen": -0.09884171932935715,
-            "logps/rejected": -0.08831708878278732,
-            "loss": 0.8677,
-            "rewards/accuracies": 0.3958333730697632,
-            "rewards/chosen": -0.1976834386587143,
-            "rewards/margins": -0.02104926109313965,
-            "rewards/rejected": -0.17663417756557465,
-            "step": 42
-        },
-        {
-            "epoch": 0.516,
-            "grad_norm": 0.9974801540374756,
-            "learning_rate": 2.8175445493671966e-07,
-            "logits/chosen": -1.3815157413482666,
-            "logits/rejected": -1.2270005941390991,
-            "logps/chosen": -0.10922063887119293,
-            "logps/rejected": -0.0900418609380722,
-            "loss": 0.8809,
-            "rewards/accuracies": 0.4583333730697632,
-            "rewards/chosen": -0.21844127774238586,
-            "rewards/margins": -0.03835754841566086,
-            "rewards/rejected": -0.1800837218761444,
-            "step": 43
-        },
-        {
-            "epoch": 0.528,
-            "grad_norm": 0.9864545464515686,
-            "learning_rate": 2.712014811188773e-07,
-            "logits/chosen": -1.3654570579528809,
-            "logits/rejected": -1.2366658449172974,
-            "logps/chosen": -0.1124953106045723,
-            "logps/rejected": -0.0964164212346077,
-            "loss": 0.8745,
-            "rewards/accuracies": 0.375,
-            "rewards/chosen": -0.2249906212091446,
-            "rewards/margins": -0.0321577824652195,
-            "rewards/rejected": -0.1928328424692154,
-            "step": 44
-        },
-        {
-            "epoch": 0.54,
-            "grad_norm": 0.9367190599441528,
-            "learning_rate": 2.606103007990371e-07,
-            "logits/chosen": -1.3880029916763306,
-            "logits/rejected": -1.2659260034561157,
-            "logps/chosen": -0.09874889254570007,
-            "logps/rejected": -0.09255427867174149,
-            "loss": 0.8628,
-            "rewards/accuracies": 0.4062500298023224,
-            "rewards/chosen": -0.19749778509140015,
-            "rewards/margins": -0.012389198876917362,
-            "rewards/rejected": -0.18510855734348297,
-            "step": 45
-        },
-        {
-            "epoch": 0.552,
-            "grad_norm": 0.9902623891830444,
-            "learning_rate": 2.5e-07,
-            "logits/chosen": -1.3836193084716797,
-            "logits/rejected": -1.2308557033538818,
-            "logps/chosen": -0.09668231755495071,
-            "logps/rejected": -0.08627666532993317,
-            "loss": 0.867,
-            "rewards/accuracies": 0.40625,
-            "rewards/chosen": -0.19336463510990143,
-            "rewards/margins": -0.020811304450035095,
-            "rewards/rejected": -0.17255333065986633,
-            "step": 46
-        },
-        {
-            "epoch": 0.564,
-            "grad_norm": 1.0337759256362915,
-            "learning_rate": 2.3938969920096296e-07,
-            "logits/chosen": -1.3623703718185425,
-            "logits/rejected": -1.2246556282043457,
-            "logps/chosen": -0.11106079071760178,
-            "logps/rejected": -0.09072640538215637,
-            "loss": 0.8814,
-            "rewards/accuracies": 0.4479166865348816,
-            "rewards/chosen": -0.22212158143520355,
-            "rewards/margins": -0.040668785572052,
-            "rewards/rejected": -0.18145281076431274,
-            "step": 47
-        },
-        {
-            "epoch": 0.576,
-            "grad_norm": 0.984899640083313,
-            "learning_rate": 2.2879851888112278e-07,
-            "logits/chosen": -1.3421802520751953,
-            "logits/rejected": -1.198030710220337,
-            "logps/chosen": -0.10126922279596329,
-            "logps/rejected": -0.09068246185779572,
-            "loss": 0.8687,
-            "rewards/accuracies": 0.3958333730697632,
-            "rewards/chosen": -0.20253844559192657,
-            "rewards/margins": -0.021173518151044846,
-            "rewards/rejected": -0.18136492371559143,
-            "step": 48
-        },
-        {
-            "epoch": 0.588,
-            "grad_norm": 1.0588122606277466,
-            "learning_rate": 2.182455450632803e-07,
-            "logits/chosen": -1.3651273250579834,
-            "logits/rejected": -1.2209078073501587,
-            "logps/chosen": -0.10214084386825562,
-            "logps/rejected": -0.09154266119003296,
-            "loss": 0.8675,
-            "rewards/accuracies": 0.4166666865348816,
-            "rewards/chosen": -0.20428168773651123,
-            "rewards/margins": -0.02119637280702591,
-            "rewards/rejected": -0.18308532238006592,
-            "step": 49
-        },
-        {
-            "epoch": 0.6,
-            "grad_norm": 1.0050568580627441,
-            "learning_rate": 2.0774979491953776e-07,
-            "logits/chosen": -1.3634512424468994,
-            "logits/rejected": -1.1983171701431274,
-            "logps/chosen": -0.09622293710708618,
-            "logps/rejected": -0.09168636053800583,
-            "loss": 0.8605,
-            "rewards/accuracies": 0.4479166865348816,
-            "rewards/chosen": -0.19244587421417236,
-            "rewards/margins": -0.009073152206838131,
-            "rewards/rejected": -0.18337272107601166,
-            "step": 50
-        },
-        {
-            "epoch": 0.612,
-            "grad_norm": 1.003779411315918,
-            "learning_rate": 1.973301825010685e-07,
-            "logits/chosen": -1.377637267112732,
-            "logits/rejected": -1.2013237476348877,
-            "logps/chosen": -0.09089501947164536,
-            "logps/rejected": -0.09312086552381516,
-            "loss": 0.8524,
-            "rewards/accuracies": 0.5520833730697632,
-            "rewards/chosen": -0.1817900389432907,
-            "rewards/margins": 0.004451685585081577,
-            "rewards/rejected": -0.1862417310476303,
-            "step": 51
-        },
-        {
-            "epoch": 0.624,
-            "grad_norm": 1.0647156238555908,
-            "learning_rate": 1.8700548465371873e-07,
-            "logits/chosen": -1.3391690254211426,
-            "logits/rejected": -1.182291030883789,
-            "logps/chosen": -0.10368049144744873,
-            "logps/rejected": -0.09800291061401367,
-            "loss": 0.8624,
-            "rewards/accuracies": 0.4895833134651184,
-            "rewards/chosen": -0.20736098289489746,
-            "rewards/margins": -0.011355183087289333,
-            "rewards/rejected": -0.19600582122802734,
-            "step": 52
-        },
-        {
-            "epoch": 0.636,
-            "grad_norm": 1.2042913436889648,
-            "learning_rate": 1.767943071808624e-07,
-            "logits/chosen": -1.3675341606140137,
-            "logits/rejected": -1.2020319700241089,
-            "logps/chosen": -0.10732070356607437,
-            "logps/rejected": -0.09919053316116333,
-            "loss": 0.871,
-            "rewards/accuracies": 0.5416666865348816,
-            "rewards/chosen": -0.21464140713214874,
-            "rewards/margins": -0.01626037061214447,
-            "rewards/rejected": -0.19838106632232666,
-            "step": 53
-        },
-        {
-            "epoch": 0.648,
-            "grad_norm": 0.9795694351196289,
-            "learning_rate": 1.667150513144856e-07,
-            "logits/chosen": -1.3203659057617188,
-            "logits/rejected": -1.1565968990325928,
-            "logps/chosen": -0.09574580192565918,
-            "logps/rejected": -0.09623640775680542,
-            "loss": 0.8547,
-            "rewards/accuracies": 0.5520833730697632,
-            "rewards/chosen": -0.19149160385131836,
-            "rewards/margins": 0.000981215387582779,
-            "rewards/rejected": -0.19247281551361084,
-            "step": 54
-        },
-        {
-            "epoch": 0.66,
-            "grad_norm": 1.0732730627059937,
-            "learning_rate": 1.5678588055492286e-07,
-            "logits/chosen": -1.2979214191436768,
-            "logits/rejected": -1.150040626525879,
-            "logps/chosen": -0.10067766904830933,
-            "logps/rejected": -0.1018737256526947,
-            "loss": 0.8545,
-            "rewards/accuracies": 0.5104166865348816,
-            "rewards/chosen": -0.20135533809661865,
-            "rewards/margins": 0.0023921187967061996,
-            "rewards/rejected": -0.2037474513053894,
-            "step": 55
-        },
-        {
-            "epoch": 0.672,
-            "grad_norm": 0.9587694406509399,
-            "learning_rate": 1.4702468793900186e-07,
-            "logits/chosen": -1.3606462478637695,
-            "logits/rejected": -1.1573679447174072,
-            "logps/chosen": -0.09973961114883423,
-            "logps/rejected": -0.10131655633449554,
-            "loss": 0.8535,
-            "rewards/accuracies": 0.5625,
-            "rewards/chosen": -0.19947922229766846,
-            "rewards/margins": 0.003153874073177576,
-            "rewards/rejected": -0.2026331126689911,
-            "step": 56
-        },
-        {
-            "epoch": 0.684,
-            "grad_norm": 1.0398699045181274,
-            "learning_rate": 1.3744906379558164e-07,
-            "logits/chosen": -1.3375906944274902,
-            "logits/rejected": -1.1461068391799927,
-            "logps/chosen": -0.10326816886663437,
-            "logps/rejected": -0.10188017785549164,
-            "loss": 0.857,
-            "rewards/accuracies": 0.4687500298023224,
-            "rewards/chosen": -0.20653633773326874,
-            "rewards/margins": -0.002775975503027439,
-            "rewards/rejected": -0.20376035571098328,
-            "step": 57
-        },
-        {
-            "epoch": 0.696,
-            "grad_norm": 1.0405802726745605,
-            "learning_rate": 1.280762640465914e-07,
-            "logits/chosen": -1.3368816375732422,
-            "logits/rejected": -1.1643409729003906,
-            "logps/chosen": -0.09207028895616531,
-            "logps/rejected": -0.09347332268953323,
-            "loss": 0.8534,
-            "rewards/accuracies": 0.5208333134651184,
-            "rewards/chosen": -0.18414057791233063,
-            "rewards/margins": 0.002806063275784254,
-            "rewards/rejected": -0.18694664537906647,
-            "step": 58
-        },
-        {
-            "epoch": 0.708,
-            "grad_norm": 1.0253372192382812,
-            "learning_rate": 1.189231791106921e-07,
-            "logits/chosen": -1.2978026866912842,
-            "logits/rejected": -1.13652765750885,
-            "logps/chosen": -0.10593652725219727,
-            "logps/rejected": -0.10165742039680481,
-            "loss": 0.8609,
-            "rewards/accuracies": 0.479166716337204,
-            "rewards/chosen": -0.21187305450439453,
-            "rewards/margins": -0.008558189496397972,
-            "rewards/rejected": -0.20331484079360962,
-            "step": 59
-        },
-        {
-            "epoch": 0.72,
-            "grad_norm": 1.2529534101486206,
-            "learning_rate": 1.1000630346560116e-07,
-            "logits/chosen": -1.3010480403900146,
-            "logits/rejected": -1.133022665977478,
-            "logps/chosen": -0.11126932501792908,
-            "logps/rejected": -0.09786901623010635,
-            "loss": 0.8738,
-            "rewards/accuracies": 0.4895833432674408,
-            "rewards/chosen": -0.22253865003585815,
-            "rewards/margins": -0.026800617575645447,
-            "rewards/rejected": -0.1957380324602127,
-            "step": 60
-        },
-        {
-            "epoch": 0.732,
-            "grad_norm": 0.9133132696151733,
-            "learning_rate": 1.0134170592392836e-07,
-            "logits/chosen": -1.3394014835357666,
-            "logits/rejected": -1.1518943309783936,
-            "logps/chosen": -0.09960196912288666,
-            "logps/rejected": -0.10244160890579224,
-            "loss": 0.8521,
-            "rewards/accuracies": 0.5104166865348816,
-            "rewards/chosen": -0.19920393824577332,
-            "rewards/margins": 0.005679287016391754,
-            "rewards/rejected": -0.20488321781158447,
-            "step": 61
-        },
-        {
-            "epoch": 0.744,
-            "grad_norm": 1.0934334993362427,
-            "learning_rate": 9.29450006760894e-08,
-            "logits/chosen": -1.3294323682785034,
-            "logits/rejected": -1.1431366205215454,
-            "logps/chosen": -0.10188900679349899,
-            "logps/rejected": -0.09653942286968231,
-            "loss": 0.8616,
-            "rewards/accuracies": 0.5,
-            "rewards/chosen": -0.20377801358699799,
-            "rewards/margins": -0.010699168778955936,
-            "rewards/rejected": -0.19307884573936462,
-            "step": 62
-        },
-        {
-            "epoch": 0.756,
-            "grad_norm": 0.9888376593589783,
-            "learning_rate": 8.483131915247967e-08,
-            "logits/chosen": -1.3347070217132568,
-            "logits/rejected": -1.167306661605835,
-            "logps/chosen": -0.10506478697061539,
-            "logps/rejected": -0.1007775291800499,
-            "loss": 0.8603,
-            "rewards/accuracies": 0.5000000596046448,
-            "rewards/chosen": -0.21012957394123077,
-            "rewards/margins": -0.008574524894356728,
-            "rewards/rejected": -0.2015550583600998,
-            "step": 63
-        },
-        {
-            "epoch": 0.768,
-            "grad_norm": 0.9486240744590759,
-            "learning_rate": 7.701528275561347e-08,
-            "logits/chosen": -1.3588067293167114,
-            "logits/rejected": -1.1604478359222412,
-            "logps/chosen": -0.0977708101272583,
-            "logps/rejected": -0.09753303229808807,
-            "loss": 0.8555,
-            "rewards/accuracies": 0.5312500596046448,
-            "rewards/chosen": -0.1955416202545166,
-            "rewards/margins": -0.0004755451809614897,
-            "rewards/rejected": -0.19506606459617615,
-            "step": 64
-        },
-        {
-            "epoch": 0.78,
-            "grad_norm": 1.033793568611145,
-            "learning_rate": 6.951097651136889e-08,
-            "logits/chosen": -1.3951979875564575,
-            "logits/rejected": -1.125361680984497,
-            "logps/chosen": -0.10307514667510986,
-            "logps/rejected": -0.1052648052573204,
-            "loss": 0.8525,
-            "rewards/accuracies": 0.4895833432674408,
-            "rewards/chosen": -0.20615029335021973,
-            "rewards/margins": 0.0043793064542114735,
-            "rewards/rejected": -0.2105296105146408,
-            "step": 65
-        },
-        {
-            "epoch": 0.792,
-            "grad_norm": 1.0308159589767456,
-            "learning_rate": 6.233192368681889e-08,
-            "logits/chosen": -1.3253390789031982,
-            "logits/rejected": -1.1783477067947388,
-            "logps/chosen": -0.09217476844787598,
-            "logps/rejected": -0.09416632354259491,
-            "loss": 0.8527,
-            "rewards/accuracies": 0.5625,
-            "rewards/chosen": -0.18434953689575195,
-            "rewards/margins": 0.003983109723776579,
-            "rewards/rejected": -0.18833264708518982,
-            "step": 66
-        },
-        {
-            "epoch": 0.804,
-            "grad_norm": 1.1081663370132446,
-            "learning_rate": 5.5491061420390174e-08,
-            "logits/chosen": -1.2740880250930786,
-            "logits/rejected": -1.152024269104004,
-            "logps/chosen": -0.10923092067241669,
-            "logps/rejected": -0.10176312923431396,
-            "loss": 0.8642,
-            "rewards/accuracies": 0.5104166865348816,
-            "rewards/chosen": -0.21846184134483337,
-            "rewards/margins": -0.01493558008223772,
-            "rewards/rejected": -0.20352625846862793,
-            "step": 67
-        },
-        {
-            "epoch": 0.816,
-            "grad_norm": 1.0150699615478516,
-            "learning_rate": 4.900071740826489e-08,
-            "logits/chosen": -1.348282814025879,
-            "logits/rejected": -1.1383142471313477,
-            "logps/chosen": -0.10064545273780823,
-            "logps/rejected": -0.10860613733530045,
-            "loss": 0.8459,
-            "rewards/accuracies": 0.6041667461395264,
-            "rewards/chosen": -0.20129090547561646,
-            "rewards/margins": 0.01592138595879078,
-            "rewards/rejected": -0.2172122746706009,
-            "step": 68
-        },
-        {
-            "epoch": 0.828,
-            "grad_norm": 1.123404860496521,
-            "learning_rate": 4.287258768903948e-08,
-            "logits/chosen": -1.3442084789276123,
-            "logits/rejected": -1.146917462348938,
-            "logps/chosen": -0.10861781984567642,
-            "logps/rejected": -0.10071661323308945,
-            "loss": 0.8688,
-            "rewards/accuracies": 0.5833333134651184,
-            "rewards/chosen": -0.21723563969135284,
-            "rewards/margins": -0.015802428126335144,
-            "rewards/rejected": -0.2014332264661789,
-            "step": 69
-        },
-        {
-            "epoch": 0.84,
-            "grad_norm": 0.9981204867362976,
-            "learning_rate": 3.7117715566672176e-08,
-            "logits/chosen": -1.3524047136306763,
-            "logits/rejected": -1.1452308893203735,
-            "logps/chosen": -0.10508691519498825,
-            "logps/rejected": -0.10403452813625336,
-            "loss": 0.8568,
-            "rewards/accuracies": 0.5729166865348816,
-            "rewards/chosen": -0.2101738303899765,
-            "rewards/margins": -0.0021047808695584536,
-            "rewards/rejected": -0.2080690562725067,
-            "step": 70
-        },
-        {
-            "epoch": 0.852,
-            "grad_norm": 1.0365066528320312,
-            "learning_rate": 3.174647170970296e-08,
-            "logits/chosen": -1.3592528104782104,
-            "logits/rejected": -1.1138660907745361,
-            "logps/chosen": -0.10330555588006973,
-            "logps/rejected": -0.1064281240105629,
-            "loss": 0.8524,
-            "rewards/accuracies": 0.5729166865348816,
-            "rewards/chosen": -0.20661111176013947,
-            "rewards/margins": 0.006245152093470097,
-            "rewards/rejected": -0.2128562480211258,
-            "step": 71
-        },
-        {
-            "epoch": 0.864,
-            "grad_norm": 1.076464295387268,
-            "learning_rate": 2.6768535462607905e-08,
-            "logits/chosen": -1.3058511018753052,
-            "logits/rejected": -1.126082420349121,
-            "logps/chosen": -0.10569944232702255,
-            "logps/rejected": -0.11077728122472763,
-            "loss": 0.8494,
-            "rewards/accuracies": 0.5,
-            "rewards/chosen": -0.2113988846540451,
-            "rewards/margins": 0.01015565823763609,
-            "rewards/rejected": -0.22155456244945526,
-            "step": 72
-        },
-        {
-            "epoch": 0.876,
-            "grad_norm": 0.9264240264892578,
-            "learning_rate": 2.2192877402966048e-08,
-            "logits/chosen": -1.3490333557128906,
-            "logits/rejected": -1.1134393215179443,
-            "logps/chosen": -0.10448520630598068,
-            "logps/rejected": -0.11283887922763824,
-            "loss": 0.8457,
-            "rewards/accuracies": 0.6145833730697632,
-            "rewards/chosen": -0.20897041261196136,
-            "rewards/margins": 0.016707373782992363,
-            "rewards/rejected": -0.2256777584552765,
-            "step": 73
-        },
-        {
-            "epoch": 0.888,
-            "grad_norm": 1.067218542098999,
-            "learning_rate": 1.8027743175872662e-08,
-            "logits/chosen": -1.345249891281128,
-            "logits/rejected": -1.1063634157180786,
-            "logps/chosen": -0.11241614818572998,
-            "logps/rejected": -0.11045221984386444,
-            "loss": 0.8579,
-            "rewards/accuracies": 0.5104166865348816,
-            "rewards/chosen": -0.22483229637145996,
-            "rewards/margins": -0.003927857149392366,
-            "rewards/rejected": -0.22090443968772888,
-            "step": 74
-        },
-        {
-            "epoch": 0.9,
-            "grad_norm": 1.0710501670837402,
-            "learning_rate": 1.4280638634728948e-08,
-            "logits/chosen": -1.3244132995605469,
-            "logits/rejected": -1.1485953330993652,
-            "logps/chosen": -0.11838357150554657,
-            "logps/rejected": -0.10926854610443115,
-            "loss": 0.8683,
-            "rewards/accuracies": 0.47916674613952637,
-            "rewards/chosen": -0.23676714301109314,
-            "rewards/margins": -0.018230034038424492,
-            "rewards/rejected": -0.2185370922088623,
-            "step": 75
-        },
-        {
-            "epoch": 0.912,
-            "grad_norm": 1.1065670251846313,
-            "learning_rate": 1.0958316315187289e-08,
-            "logits/chosen": -1.317086100578308,
-            "logits/rejected": -1.1413421630859375,
-            "logps/chosen": -0.10257872194051743,
-            "logps/rejected": -0.10372138023376465,
-            "loss": 0.8542,
-            "rewards/accuracies": 0.5416666269302368,
-            "rewards/chosen": -0.20515744388103485,
-            "rewards/margins": 0.0022853193804621696,
-            "rewards/rejected": -0.2074427604675293,
-            "step": 76
-        },
-        {
-            "epoch": 0.924,
-            "grad_norm": 1.0825769901275635,
-            "learning_rate": 8.066763266625282e-09,
-            "logits/chosen": -1.3573386669158936,
-            "logits/rejected": -1.109717607498169,
-            "logps/chosen": -0.10758916288614273,
-            "logps/rejected": -0.10599493980407715,
-            "loss": 0.8597,
-            "rewards/accuracies": 0.6250000596046448,
-            "rewards/chosen": -0.21517832577228546,
-            "rewards/margins": -0.0031884238123893738,
-            "rewards/rejected": -0.2119898796081543,
-            "step": 77
-        },
-        {
-            "epoch": 0.936,
-            "grad_norm": 0.9156083464622498,
-            "learning_rate": 5.611190263078463e-09,
-            "logits/chosen": -1.309991717338562,
-            "logits/rejected": -1.1210401058197021,
-            "logps/chosen": -0.0973024070262909,
-            "logps/rejected": -0.09729278087615967,
-            "loss": 0.8551,
-            "rewards/accuracies": 0.4583333432674408,
-            "rewards/chosen": -0.1946048140525818,
-            "rewards/margins": -1.9263941794633865e-05,
-            "rewards/rejected": -0.19458556175231934,
-            "step": 78
-        },
-        {
-            "epoch": 0.948,
-            "grad_norm": 0.9994163513183594,
-            "learning_rate": 3.5960224130728858e-09,
-            "logits/chosen": -1.3023028373718262,
-            "logits/rejected": -1.1124149560928345,
-            "logps/chosen": -0.09809858351945877,
-            "logps/rejected": -0.1080770194530487,
-            "loss": 0.8436,
-            "rewards/accuracies": 0.6354166865348816,
-            "rewards/chosen": -0.19619716703891754,
-            "rewards/margins": 0.01995689421892166,
-            "rewards/rejected": -0.2161540389060974,
-            "step": 79
-        },
-        {
-            "epoch": 0.96,
-            "grad_norm": 0.9938739538192749,
-            "learning_rate": 2.0248911852807917e-09,
-            "logits/chosen": -1.3484057188034058,
-            "logits/rejected": -1.0871906280517578,
-            "logps/chosen": -0.11684219539165497,
-            "logps/rejected": -0.11286689341068268,
-            "loss": 0.8601,
-            "rewards/accuracies": 0.5104166865348816,
-            "rewards/chosen": -0.23368439078330994,
-            "rewards/margins": -0.007950600236654282,
-            "rewards/rejected": -0.22573378682136536,
-            "step": 80
-        },
-        {
-            "epoch": 0.972,
-            "grad_norm": 0.9481876492500305,
-            "learning_rate": 9.006278643683696e-10,
-            "logits/chosen": -1.3169337511062622,
-            "logits/rejected": -1.1376291513442993,
-            "logps/chosen": -0.09936561435461044,
-            "logps/rejected": -0.10312428325414658,
-            "loss": 0.8508,
-            "rewards/accuracies": 0.5833333730697632,
-            "rewards/chosen": -0.1987312287092209,
-            "rewards/margins": 0.007517362013459206,
-            "rewards/rejected": -0.20624856650829315,
-            "step": 81
-        },
-        {
-            "epoch": 0.984,
-            "grad_norm": 1.0620635747909546,
-            "learning_rate": 2.2525844882964606e-10,
-            "logits/chosen": -1.3467659950256348,
-            "logits/rejected": -1.1462361812591553,
-            "logps/chosen": -0.10522940754890442,
-            "logps/rejected": -0.10153805464506149,
-            "loss": 0.8594,
-            "rewards/accuracies": 0.4687500298023224,
-            "rewards/chosen": -0.21045881509780884,
-            "rewards/margins": -0.007382689975202084,
-            "rewards/rejected": -0.20307610929012299,
-            "step": 82
-        },
-        {
-            "epoch": 0.996,
-            "grad_norm": 1.0221400260925293,
-            "learning_rate": 0.0,
-            "logits/chosen": -1.313194751739502,
-            "logits/rejected": -1.164141297340393,
-            "logps/chosen": -0.10121379047632217,
-            "logps/rejected": -0.1018705815076828,
-            "loss": 0.8549,
-            "rewards/accuracies": 0.5416666865348816,
-            "rewards/chosen": -0.20242758095264435,
-            "rewards/margins": 0.0013135506305843592,
-            "rewards/rejected": -0.2037411630153656,
-            "step": 83
-        }
-    ],
-    "logging_steps": 1,
-    "max_steps": 83,
-    "num_input_tokens_seen": 0,
-    "num_train_epochs": 1,
-    "save_steps": 100,
-    "stateful_callbacks": {
-        "TrainerControl": {
-            "args": {
-                "should_epoch_stop": false,
-                "should_evaluate": false,
-                "should_log": false,
-                "should_save": true,
-                "should_training_stop": true
-            },
-            "attributes": {}
-        }
-    },
-    "total_flos": 305651334512640.0,
-    "train_batch_size": 1,
-    "trial_name": null,
-    "trial_params": null
-}

+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.996,
+  "eval_steps": 500,
+  "global_step": 83,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.012,
+      "grad_norm": 0.8640882968902588,
+      "learning_rate": 5.555555555555555e-08,
+      "logits/chosen": -1.495650291442871,
+      "logits/rejected": -1.3535889387130737,
+      "logps/chosen": -0.10173828899860382,
+      "logps/rejected": -0.08792766183614731,
+      "loss": 0.8717,
+      "rewards/accuracies": 0.3333333730697632,
+      "rewards/chosen": -0.20347657799720764,
+      "rewards/margins": -0.027621246874332428,
+      "rewards/rejected": -0.17585532367229462,
+      "step": 1
+    },
+    {
+      "epoch": 0.024,
+      "grad_norm": 0.979682981967926,
+      "learning_rate": 1.111111111111111e-07,
+      "logits/chosen": -1.443913221359253,
+      "logits/rejected": -1.3641024827957153,
+      "logps/chosen": -0.10752908140420914,
+      "logps/rejected": -0.08240076899528503,
+      "loss": 0.8846,
+      "rewards/accuracies": 0.2187500149011612,
+      "rewards/chosen": -0.21505816280841827,
+      "rewards/margins": -0.0502566322684288,
+      "rewards/rejected": -0.16480153799057007,
+      "step": 2
+    },
+    {
+      "epoch": 0.036,
+      "grad_norm": 0.9309831857681274,
+      "learning_rate": 1.6666666666666665e-07,
+      "logits/chosen": -1.4732969999313354,
+      "logits/rejected": -1.376213550567627,
+      "logps/chosen": -0.10481980443000793,
+      "logps/rejected": -0.07933054119348526,
+      "loss": 0.8848,
+      "rewards/accuracies": 0.28125,
+      "rewards/chosen": -0.20963960886001587,
+      "rewards/margins": -0.050978537648916245,
+      "rewards/rejected": -0.15866108238697052,
+      "step": 3
+    },
+    {
+      "epoch": 0.048,
+      "grad_norm": 0.9682196974754333,
+      "learning_rate": 2.222222222222222e-07,
+      "logits/chosen": -1.5285106897354126,
+      "logits/rejected": -1.3950246572494507,
+      "logps/chosen": -0.11225953698158264,
+      "logps/rejected": -0.08694332838058472,
+      "loss": 0.8849,
+      "rewards/accuracies": 0.3229166865348816,
+      "rewards/chosen": -0.22451907396316528,
+      "rewards/margins": -0.05063238739967346,
+      "rewards/rejected": -0.17388665676116943,
+      "step": 4
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 1.0820338726043701,
+      "learning_rate": 2.7777777777777776e-07,
+      "logits/chosen": -1.494052767753601,
+      "logits/rejected": -1.3356661796569824,
+      "logps/chosen": -0.12428013980388641,
+      "logps/rejected": -0.09185128659009933,
+      "loss": 0.8968,
+      "rewards/accuracies": 0.322916716337204,
+      "rewards/chosen": -0.24856027960777283,
+      "rewards/margins": -0.06485769897699356,
+      "rewards/rejected": -0.18370257318019867,
+      "step": 5
+    },
+    {
+      "epoch": 0.072,
+      "grad_norm": 1.0573477745056152,
+      "learning_rate": 3.333333333333333e-07,
+      "logits/chosen": -1.4825295209884644,
+      "logits/rejected": -1.3897716999053955,
+      "logps/chosen": -0.11380095779895782,
+      "logps/rejected": -0.08133920282125473,
+      "loss": 0.8938,
+      "rewards/accuracies": 0.3020833432674408,
+      "rewards/chosen": -0.22760191559791565,
+      "rewards/margins": -0.06492353230714798,
+      "rewards/rejected": -0.16267840564250946,
+      "step": 6
+    },
+    {
+      "epoch": 0.084,
+      "grad_norm": 0.9294664263725281,
+      "learning_rate": 3.888888888888889e-07,
+      "logits/chosen": -1.5021216869354248,
+      "logits/rejected": -1.3295238018035889,
+      "logps/chosen": -0.099081851541996,
+      "logps/rejected": -0.08119318634271622,
+      "loss": 0.8761,
+      "rewards/accuracies": 0.2916666567325592,
+      "rewards/chosen": -0.198163703083992,
+      "rewards/margins": -0.03577733412384987,
+      "rewards/rejected": -0.16238637268543243,
+      "step": 7
+    },
+    {
+      "epoch": 0.096,
+      "grad_norm": 1.0329580307006836,
+      "learning_rate": 4.444444444444444e-07,
+      "logits/chosen": -1.5048794746398926,
+      "logits/rejected": -1.377795934677124,
+      "logps/chosen": -0.12124787271022797,
+      "logps/rejected": -0.08464659005403519,
+      "loss": 0.9015,
+      "rewards/accuracies": 0.3125,
+      "rewards/chosen": -0.24249574542045593,
+      "rewards/margins": -0.07320256531238556,
+      "rewards/rejected": -0.16929318010807037,
+      "step": 8
+    },
+    {
+      "epoch": 0.108,
+      "grad_norm": 0.9767515063285828,
+      "learning_rate": 5e-07,
+      "logits/chosen": -1.5186768770217896,
+      "logits/rejected": -1.3877286911010742,
+      "logps/chosen": -0.10737170279026031,
+      "logps/rejected": -0.0823729932308197,
+      "loss": 0.8847,
+      "rewards/accuracies": 0.3020833432674408,
+      "rewards/chosen": -0.21474340558052063,
+      "rewards/margins": -0.04999742656946182,
+      "rewards/rejected": -0.1647459864616394,
+      "step": 9
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.9653743505477905,
+      "learning_rate": 4.997747415511704e-07,
+      "logits/chosen": -1.4915320873260498,
+      "logits/rejected": -1.3450301885604858,
+      "logps/chosen": -0.10813666880130768,
+      "logps/rejected": -0.08507634699344635,
+      "loss": 0.8822,
+      "rewards/accuracies": 0.2291666716337204,
+      "rewards/chosen": -0.21627333760261536,
+      "rewards/margins": -0.04612065851688385,
+      "rewards/rejected": -0.1701526939868927,
+      "step": 10
+    },
+    {
+      "epoch": 0.132,
+      "grad_norm": 0.9533908367156982,
+      "learning_rate": 4.990993721356315e-07,
+      "logits/chosen": -1.4848368167877197,
+      "logits/rejected": -1.3539990186691284,
+      "logps/chosen": -0.11860796809196472,
+      "logps/rejected": -0.08763512969017029,
+      "loss": 0.8931,
+      "rewards/accuracies": 0.25,
+      "rewards/chosen": -0.23721593618392944,
+      "rewards/margins": -0.061945684254169464,
+      "rewards/rejected": -0.17527025938034058,
+      "step": 11
+    },
+    {
+      "epoch": 0.144,
+      "grad_norm": 0.9020726084709167,
+      "learning_rate": 4.979751088147191e-07,
+      "logits/chosen": -1.4897594451904297,
+      "logits/rejected": -1.3646252155303955,
+      "logps/chosen": -0.09957993775606155,
+      "logps/rejected": -0.08218874782323837,
+      "loss": 0.8759,
+      "rewards/accuracies": 0.3958333432674408,
+      "rewards/chosen": -0.1991598755121231,
+      "rewards/margins": -0.034782394766807556,
+      "rewards/rejected": -0.16437749564647675,
+      "step": 12
+    },
+    {
+      "epoch": 0.156,
+      "grad_norm": 1.0148029327392578,
+      "learning_rate": 4.964039775869271e-07,
+      "logits/chosen": -1.4809292554855347,
+      "logits/rejected": -1.3655236959457397,
+      "logps/chosen": -0.11366236209869385,
+      "logps/rejected": -0.0875682383775711,
+      "loss": 0.8868,
+      "rewards/accuracies": 0.3645833432674408,
+      "rewards/chosen": -0.2273247241973877,
+      "rewards/margins": -0.05218825116753578,
+      "rewards/rejected": -0.1751364767551422,
+      "step": 13
+    },
+    {
+      "epoch": 0.168,
+      "grad_norm": 1.334598422050476,
+      "learning_rate": 4.943888097369216e-07,
+      "logits/chosen": -1.4804601669311523,
+      "logits/rejected": -1.3518431186676025,
+      "logps/chosen": -0.13359910249710083,
+      "logps/rejected": -0.08674132823944092,
+      "loss": 0.9236,
+      "rewards/accuracies": 0.2395833432674408,
+      "rewards/chosen": -0.26719820499420166,
+      "rewards/margins": -0.09371551126241684,
+      "rewards/rejected": -0.17348265647888184,
+      "step": 14
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.9834568500518799,
+      "learning_rate": 4.919332367333748e-07,
+      "logits/chosen": -1.4974645376205444,
+      "logits/rejected": -1.3400018215179443,
+      "logps/chosen": -0.10907550156116486,
+      "logps/rejected": -0.08122590184211731,
+      "loss": 0.8879,
+      "rewards/accuracies": 0.21875,
+      "rewards/chosen": -0.2181510031223297,
+      "rewards/margins": -0.05569921433925629,
+      "rewards/rejected": -0.16245180368423462,
+      "step": 15
+    },
+    {
+      "epoch": 0.192,
+      "grad_norm": 0.9361069798469543,
+      "learning_rate": 4.890416836848127e-07,
+      "logits/chosen": -1.4610369205474854,
+      "logits/rejected": -1.369497299194336,
+      "logps/chosen": -0.10335300862789154,
+      "logps/rejected": -0.08525583893060684,
+      "loss": 0.8765,
+      "rewards/accuracies": 0.3541666865348816,
+      "rewards/chosen": -0.20670601725578308,
+      "rewards/margins": -0.03619435429573059,
+      "rewards/rejected": -0.17051167786121368,
+      "step": 16
+    },
+    {
+      "epoch": 0.204,
+      "grad_norm": 0.910740077495575,
+      "learning_rate": 4.85719361365271e-07,
+      "logits/chosen": -1.5087649822235107,
+      "logits/rejected": -1.3652905225753784,
+      "logps/chosen": -0.09960392862558365,
+      "logps/rejected": -0.08507402241230011,
+      "loss": 0.8719,
+      "rewards/accuracies": 0.3541666865348816,
+      "rewards/chosen": -0.1992078572511673,
+      "rewards/margins": -0.029059793800115585,
+      "rewards/rejected": -0.17014804482460022,
+      "step": 17
+    },
+    {
+      "epoch": 0.216,
+      "grad_norm": 0.9233217239379883,
+      "learning_rate": 4.819722568241273e-07,
+      "logits/chosen": -1.4695565700531006,
+      "logits/rejected": -1.336183786392212,
+      "logps/chosen": -0.10343371331691742,
+      "logps/rejected": -0.0857667475938797,
+      "loss": 0.8769,
+      "rewards/accuracies": 0.4166666865348816,
+      "rewards/chosen": -0.20686742663383484,
+      "rewards/margins": -0.03533393144607544,
+      "rewards/rejected": -0.1715334951877594,
+      "step": 18
+    },
+    {
+      "epoch": 0.228,
+      "grad_norm": 0.8909419775009155,
+      "learning_rate": 4.778071225970339e-07,
+      "logits/chosen": -1.4759807586669922,
+      "logits/rejected": -1.3495370149612427,
+      "logps/chosen": -0.10235996544361115,
+      "logps/rejected": -0.0851350948214531,
+      "loss": 0.8757,
+      "rewards/accuracies": 0.3229166567325592,
+      "rewards/chosen": -0.2047199308872223,
+      "rewards/margins": -0.03444972261786461,
+      "rewards/rejected": -0.1702701896429062,
+      "step": 19
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.9479613304138184,
+      "learning_rate": 4.732314645373921e-07,
+      "logits/chosen": -1.4673627614974976,
+      "logits/rejected": -1.3651877641677856,
+      "logps/chosen": -0.10552840679883957,
+      "logps/rejected": -0.07911674678325653,
+      "loss": 0.886,
+      "rewards/accuracies": 0.28125,
+      "rewards/chosen": -0.21105681359767914,
+      "rewards/margins": -0.05282333493232727,
+      "rewards/rejected": -0.15823349356651306,
+      "step": 20
+    },
+    {
+      "epoch": 0.252,
+      "grad_norm": 0.854266345500946,
+      "learning_rate": 4.68253528290297e-07,
+      "logits/chosen": -1.4877924919128418,
+      "logits/rejected": -1.324515461921692,
+      "logps/chosen": -0.1002732366323471,
+      "logps/rejected": -0.09079625457525253,
+      "loss": 0.8664,
+      "rewards/accuracies": 0.3645833432674408,
+      "rewards/chosen": -0.2005464732646942,
+      "rewards/margins": -0.018953965976834297,
+      "rewards/rejected": -0.18159250915050507,
+      "step": 21
+    },
+    {
+      "epoch": 0.264,
+      "grad_norm": 0.9627916812896729,
+      "learning_rate": 4.6288228443332776e-07,
+      "logits/chosen": -1.466447114944458,
+      "logits/rejected": -1.3485172986984253,
+      "logps/chosen": -0.10300128906965256,
+      "logps/rejected": -0.08484764397144318,
+      "loss": 0.8763,
+      "rewards/accuracies": 0.3333333730697632,
+      "rewards/chosen": -0.20600257813930511,
+      "rewards/margins": -0.03630730137228966,
+      "rewards/rejected": -0.16969528794288635,
+      "step": 22
+    },
+    {
+      "epoch": 0.276,
+      "grad_norm": 0.8689938187599182,
+      "learning_rate": 4.571274123109605e-07,
+      "logits/chosen": -1.4124252796173096,
+      "logits/rejected": -1.3001394271850586,
+      "logps/chosen": -0.10588695108890533,
+      "logps/rejected": -0.08438257873058319,
+      "loss": 0.8809,
+      "rewards/accuracies": 0.2395833283662796,
+      "rewards/chosen": -0.21177390217781067,
+      "rewards/margins": -0.043008752167224884,
+      "rewards/rejected": -0.16876515746116638,
+      "step": 23
+    },
+    {
+      "epoch": 0.288,
+      "grad_norm": 0.9801927208900452,
+      "learning_rate": 4.5099928259173514e-07,
+      "logits/chosen": -1.4522674083709717,
+      "logits/rejected": -1.3461543321609497,
+      "logps/chosen": -0.1134054884314537,
+      "logps/rejected": -0.08870639652013779,
+      "loss": 0.886,
+      "rewards/accuracies": 0.2499999850988388,
+      "rewards/chosen": -0.2268109768629074,
+      "rewards/margins": -0.049398161470890045,
+      "rewards/rejected": -0.17741279304027557,
+      "step": 24
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.9185568690299988,
+      "learning_rate": 4.4450893857960984e-07,
+      "logits/chosen": -1.4601349830627441,
+      "logits/rejected": -1.273384928703308,
+      "logps/chosen": -0.10715562850236893,
+      "logps/rejected": -0.09089934825897217,
+      "loss": 0.8746,
+      "rewards/accuracies": 0.375,
+      "rewards/chosen": -0.21431125700473785,
+      "rewards/margins": -0.03251257538795471,
+      "rewards/rejected": -0.18179869651794434,
+      "step": 25
+    },
+    {
+      "epoch": 0.312,
+      "grad_norm": 0.8864983320236206,
+      "learning_rate": 4.3766807631318105e-07,
+      "logits/chosen": -1.438947081565857,
+      "logits/rejected": -1.3086212873458862,
+      "logps/chosen": -0.10850708931684494,
+      "logps/rejected": -0.0892128050327301,
+      "loss": 0.8781,
+      "rewards/accuracies": 0.3229166865348816,
+      "rewards/chosen": -0.21701417863368988,
+      "rewards/margins": -0.038588590919971466,
+      "rewards/rejected": -0.1784256100654602,
+      "step": 26
+    },
+    {
+      "epoch": 0.324,
+      "grad_norm": 0.9518214464187622,
+      "learning_rate": 4.3048902348863106e-07,
+      "logits/chosen": -1.4363845586776733,
+      "logits/rejected": -1.3129128217697144,
+      "logps/chosen": -0.10809854418039322,
+      "logps/rejected": -0.08249183744192123,
+      "loss": 0.885,
+      "rewards/accuracies": 0.2812500298023224,
+      "rewards/chosen": -0.21619708836078644,
+      "rewards/margins": -0.05121342092752457,
+      "rewards/rejected": -0.16498367488384247,
+      "step": 27
+    },
+    {
+      "epoch": 0.336,
+      "grad_norm": 0.9291889667510986,
+      "learning_rate": 4.2298471724438653e-07,
+      "logits/chosen": -1.4329116344451904,
+      "logits/rejected": -1.3157219886779785,
+      "logps/chosen": -0.10641689598560333,
+      "logps/rejected": -0.0836237370967865,
+      "loss": 0.8825,
+      "rewards/accuracies": 0.3541666567325592,
+      "rewards/chosen": -0.21283379197120667,
+      "rewards/margins": -0.04558631405234337,
+      "rewards/rejected": -0.167247474193573,
+      "step": 28
+    },
+    {
+      "epoch": 0.348,
+      "grad_norm": 0.8883063197135925,
+      "learning_rate": 4.151686808475203e-07,
+      "logits/chosen": -1.4277637004852295,
+      "logits/rejected": -1.2657394409179688,
+      "logps/chosen": -0.1044892817735672,
+      "logps/rejected": -0.08842873573303223,
+      "loss": 0.8741,
+      "rewards/accuracies": 0.34375,
+      "rewards/chosen": -0.2089785635471344,
+      "rewards/margins": -0.03212107717990875,
+      "rewards/rejected": -0.17685747146606445,
+      "step": 29
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.9316148161888123,
+      "learning_rate": 4.070549993239106e-07,
+      "logits/chosen": -1.3881316184997559,
+      "logits/rejected": -1.2497737407684326,
+      "logps/chosen": -0.11017285287380219,
+      "logps/rejected": -0.0868031308054924,
+      "loss": 0.8842,
+      "rewards/accuracies": 0.3645833432674408,
+      "rewards/chosen": -0.22034570574760437,
+      "rewards/margins": -0.046739429235458374,
+      "rewards/rejected": -0.1736062616109848,
+      "step": 30
+    },
+    {
+      "epoch": 0.372,
+      "grad_norm": 0.8768561482429504,
+      "learning_rate": 3.9865829407607166e-07,
+      "logits/chosen": -1.4147083759307861,
+      "logits/rejected": -1.2981321811676025,
+      "logps/chosen": -0.10104362666606903,
+      "logps/rejected": -0.09066756814718246,
+      "loss": 0.8673,
+      "rewards/accuracies": 0.3437500298023224,
+      "rewards/chosen": -0.20208725333213806,
+      "rewards/margins": -0.020752109587192535,
+      "rewards/rejected": -0.18133513629436493,
+      "step": 31
+    },
+    {
+      "epoch": 0.384,
+      "grad_norm": 1.0077239274978638,
+      "learning_rate": 3.8999369653439883e-07,
+      "logits/chosen": -1.4155137538909912,
+      "logits/rejected": -1.2743993997573853,
+      "logps/chosen": -0.10902610421180725,
+      "logps/rejected": -0.0863211378455162,
+      "loss": 0.8827,
+      "rewards/accuracies": 0.3020833432674408,
+      "rewards/chosen": -0.2180522084236145,
+      "rewards/margins": -0.0454099103808403,
+      "rewards/rejected": -0.1726422756910324,
+      "step": 32
+    },
+    {
+      "epoch": 0.396,
+      "grad_norm": 0.9207706451416016,
+      "learning_rate": 3.810768208893079e-07,
+      "logits/chosen": -1.3758422136306763,
+      "logits/rejected": -1.2915358543395996,
+      "logps/chosen": -0.1034766435623169,
+      "logps/rejected": -0.07977995276451111,
+      "loss": 0.883,
+      "rewards/accuracies": 0.3020833432674408,
+      "rewards/chosen": -0.2069532871246338,
+      "rewards/margins": -0.047393374145030975,
+      "rewards/rejected": -0.15955990552902222,
+      "step": 33
+    },
+    {
+      "epoch": 0.408,
+      "grad_norm": 0.9844294190406799,
+      "learning_rate": 3.7192373595340865e-07,
+      "logits/chosen": -1.442295789718628,
+      "logits/rejected": -1.2863086462020874,
+      "logps/chosen": -0.10710459202528,
+      "logps/rejected": -0.08674141019582748,
+      "loss": 0.8796,
+      "rewards/accuracies": 0.3020833432674408,
+      "rewards/chosen": -0.21420918405056,
+      "rewards/margins": -0.04072638228535652,
+      "rewards/rejected": -0.17348282039165497,
+      "step": 34
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.9682800769805908,
+      "learning_rate": 3.625509362044183e-07,
+      "logits/chosen": -1.3701705932617188,
+      "logits/rejected": -1.2656924724578857,
+      "logps/chosen": -0.11000403016805649,
+      "logps/rejected": -0.08980907499790192,
+      "loss": 0.8803,
+      "rewards/accuracies": 0.3854166567325592,
+      "rewards/chosen": -0.22000806033611298,
+      "rewards/margins": -0.04038992151618004,
+      "rewards/rejected": -0.17961814999580383,
+      "step": 35
+    },
+    {
+      "epoch": 0.432,
+      "grad_norm": 0.9004039168357849,
+      "learning_rate": 3.529753120609982e-07,
+      "logits/chosen": -1.4099225997924805,
+      "logits/rejected": -1.252682089805603,
+      "logps/chosen": -0.09707480669021606,
+      "logps/rejected": -0.08859608322381973,
+      "loss": 0.8653,
+      "rewards/accuracies": 0.4479166865348816,
+      "rewards/chosen": -0.19414961338043213,
+      "rewards/margins": -0.016957445070147514,
+      "rewards/rejected": -0.17719216644763947,
+      "step": 36
+    },
+    {
+      "epoch": 0.444,
+      "grad_norm": 0.9568714499473572,
+      "learning_rate": 3.4321411944507714e-07,
+      "logits/chosen": -1.3612836599349976,
+      "logits/rejected": -1.2257184982299805,
+      "logps/chosen": -0.10623904317617416,
+      "logps/rejected": -0.09100136905908585,
+      "loss": 0.8734,
+      "rewards/accuracies": 0.4166666567325592,
+      "rewards/chosen": -0.21247808635234833,
+      "rewards/margins": -0.030475351959466934,
+      "rewards/rejected": -0.1820027381181717,
+      "step": 37
+    },
+    {
+      "epoch": 0.456,
+      "grad_norm": 0.8658465147018433,
+      "learning_rate": 3.332849486855144e-07,
+      "logits/chosen": -1.381141185760498,
+      "logits/rejected": -1.2495661973953247,
+      "logps/chosen": -0.09047922492027283,
+      "logps/rejected": -0.07975338399410248,
+      "loss": 0.8676,
+      "rewards/accuracies": 0.3958333432674408,
+      "rewards/chosen": -0.18095844984054565,
+      "rewards/margins": -0.021451696753501892,
+      "rewards/rejected": -0.15950676798820496,
+      "step": 38
+    },
+    {
+      "epoch": 0.468,
+      "grad_norm": 0.9716626405715942,
+      "learning_rate": 3.2320569281913754e-07,
+      "logits/chosen": -1.3790628910064697,
+      "logits/rejected": -1.251636028289795,
+      "logps/chosen": -0.10486049205064774,
+      "logps/rejected": -0.08277000486850739,
+      "loss": 0.883,
+      "rewards/accuracies": 0.3437500298023224,
+      "rewards/chosen": -0.20972098410129547,
+      "rewards/margins": -0.0441809706389904,
+      "rewards/rejected": -0.16554000973701477,
+      "step": 39
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.948078989982605,
+      "learning_rate": 3.129945153462813e-07,
+      "logits/chosen": -1.3937729597091675,
+      "logits/rejected": -1.2695637941360474,
+      "logps/chosen": -0.09631801396608353,
+      "logps/rejected": -0.08585619181394577,
+      "loss": 0.8673,
+      "rewards/accuracies": 0.3854166865348816,
+      "rewards/chosen": -0.19263602793216705,
+      "rewards/margins": -0.020923633128404617,
+      "rewards/rejected": -0.17171238362789154,
+      "step": 40
+    },
+    {
+      "epoch": 0.492,
+      "grad_norm": 0.9445565342903137,
+      "learning_rate": 3.0266981749893154e-07,
+      "logits/chosen": -1.3988057374954224,
+      "logits/rejected": -1.2703980207443237,
+      "logps/chosen": -0.09746824949979782,
+      "logps/rejected": -0.0910995602607727,
+      "loss": 0.8624,
+      "rewards/accuracies": 0.40625,
+      "rewards/chosen": -0.19493649899959564,
+      "rewards/margins": -0.012737366370856762,
+      "rewards/rejected": -0.1821991205215454,
+      "step": 41
+    },
+    {
+      "epoch": 0.504,
+      "grad_norm": 1.080769419670105,
+      "learning_rate": 2.922502050804623e-07,
+      "logits/chosen": -1.4094312191009521,
+      "logits/rejected": -1.247127890586853,
+      "logps/chosen": -0.09884171932935715,
+      "logps/rejected": -0.08831708878278732,
+      "loss": 0.8677,
+      "rewards/accuracies": 0.3958333730697632,
+      "rewards/chosen": -0.1976834386587143,
+      "rewards/margins": -0.02104926109313965,
+      "rewards/rejected": -0.17663417756557465,
+      "step": 42
+    },
+    {
+      "epoch": 0.516,
+      "grad_norm": 0.9974801540374756,
+      "learning_rate": 2.8175445493671966e-07,
+      "logits/chosen": -1.3815157413482666,
+      "logits/rejected": -1.2270005941390991,
+      "logps/chosen": -0.10922063887119293,
+      "logps/rejected": -0.0900418609380722,
+      "loss": 0.8809,
+      "rewards/accuracies": 0.4583333730697632,
+      "rewards/chosen": -0.21844127774238586,
+      "rewards/margins": -0.03835754841566086,
+      "rewards/rejected": -0.1800837218761444,
+      "step": 43
+    },
+    {
+      "epoch": 0.528,
+      "grad_norm": 0.9864545464515686,
+      "learning_rate": 2.712014811188773e-07,
+      "logits/chosen": -1.3654570579528809,
+      "logits/rejected": -1.2366658449172974,
+      "logps/chosen": -0.1124953106045723,
+      "logps/rejected": -0.0964164212346077,
+      "loss": 0.8745,
+      "rewards/accuracies": 0.375,
+      "rewards/chosen": -0.2249906212091446,
+      "rewards/margins": -0.0321577824652195,
+      "rewards/rejected": -0.1928328424692154,
+      "step": 44
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.9367190599441528,
+      "learning_rate": 2.606103007990371e-07,
+      "logits/chosen": -1.3880029916763306,
+      "logits/rejected": -1.2659260034561157,
+      "logps/chosen": -0.09874889254570007,
+      "logps/rejected": -0.09255427867174149,
+      "loss": 0.8628,
+      "rewards/accuracies": 0.4062500298023224,
+      "rewards/chosen": -0.19749778509140015,
+      "rewards/margins": -0.012389198876917362,
+      "rewards/rejected": -0.18510855734348297,
+      "step": 45
+    },
+    {
+      "epoch": 0.552,
+      "grad_norm": 0.9902623891830444,
+      "learning_rate": 2.5e-07,
+      "logits/chosen": -1.3836193084716797,
+      "logits/rejected": -1.2308557033538818,
+      "logps/chosen": -0.09668231755495071,
+      "logps/rejected": -0.08627666532993317,
+      "loss": 0.867,
+      "rewards/accuracies": 0.40625,
+      "rewards/chosen": -0.19336463510990143,
+      "rewards/margins": -0.020811304450035095,
+      "rewards/rejected": -0.17255333065986633,
+      "step": 46
+    },
+    {
+      "epoch": 0.564,
+      "grad_norm": 1.0337759256362915,
+      "learning_rate": 2.3938969920096296e-07,
+      "logits/chosen": -1.3623703718185425,
+      "logits/rejected": -1.2246556282043457,
+      "logps/chosen": -0.11106079071760178,
+      "logps/rejected": -0.09072640538215637,
+      "loss": 0.8814,
+      "rewards/accuracies": 0.4479166865348816,
+      "rewards/chosen": -0.22212158143520355,
+      "rewards/margins": -0.040668785572052,
+      "rewards/rejected": -0.18145281076431274,
+      "step": 47
+    },
+    {
+      "epoch": 0.576,
+      "grad_norm": 0.984899640083313,
+      "learning_rate": 2.2879851888112278e-07,
+      "logits/chosen": -1.3421802520751953,
+      "logits/rejected": -1.198030710220337,
+      "logps/chosen": -0.10126922279596329,
+      "logps/rejected": -0.09068246185779572,
+      "loss": 0.8687,
+      "rewards/accuracies": 0.3958333730697632,
+      "rewards/chosen": -0.20253844559192657,
+      "rewards/margins": -0.021173518151044846,
+      "rewards/rejected": -0.18136492371559143,
+      "step": 48
+    },
+    {
+      "epoch": 0.588,
+      "grad_norm": 1.0588122606277466,
+      "learning_rate": 2.182455450632803e-07,
+      "logits/chosen": -1.3651273250579834,
+      "logits/rejected": -1.2209078073501587,
+      "logps/chosen": -0.10214084386825562,
+      "logps/rejected": -0.09154266119003296,
+      "loss": 0.8675,
+      "rewards/accuracies": 0.4166666865348816,
+      "rewards/chosen": -0.20428168773651123,
+      "rewards/margins": -0.02119637280702591,
+      "rewards/rejected": -0.18308532238006592,
+      "step": 49
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 1.0050568580627441,
+      "learning_rate": 2.0774979491953776e-07,
+      "logits/chosen": -1.3634512424468994,
+      "logits/rejected": -1.1983171701431274,
+      "logps/chosen": -0.09622293710708618,
+      "logps/rejected": -0.09168636053800583,
+      "loss": 0.8605,
+      "rewards/accuracies": 0.4479166865348816,
+      "rewards/chosen": -0.19244587421417236,
+      "rewards/margins": -0.009073152206838131,
+      "rewards/rejected": -0.18337272107601166,
+      "step": 50
+    },
+    {
+      "epoch": 0.612,
+      "grad_norm": 1.003779411315918,
+      "learning_rate": 1.973301825010685e-07,
+      "logits/chosen": -1.377637267112732,
+      "logits/rejected": -1.2013237476348877,
+      "logps/chosen": -0.09089501947164536,
+      "logps/rejected": -0.09312086552381516,
+      "loss": 0.8524,
+      "rewards/accuracies": 0.5520833730697632,
+      "rewards/chosen": -0.1817900389432907,
+      "rewards/margins": 0.004451685585081577,
+      "rewards/rejected": -0.1862417310476303,
+      "step": 51
+    },
+    {
+      "epoch": 0.624,
+      "grad_norm": 1.0647156238555908,
+      "learning_rate": 1.8700548465371873e-07,
+      "logits/chosen": -1.3391690254211426,
+      "logits/rejected": -1.182291030883789,
+      "logps/chosen": -0.10368049144744873,
+      "logps/rejected": -0.09800291061401367,
+      "loss": 0.8624,
+      "rewards/accuracies": 0.4895833134651184,
+      "rewards/chosen": -0.20736098289489746,
+      "rewards/margins": -0.011355183087289333,
+      "rewards/rejected": -0.19600582122802734,
+      "step": 52
+    },
+    {
+      "epoch": 0.636,
+      "grad_norm": 1.2042913436889648,
+      "learning_rate": 1.767943071808624e-07,
+      "logits/chosen": -1.3675341606140137,
+      "logits/rejected": -1.2020319700241089,
+      "logps/chosen": -0.10732070356607437,
+      "logps/rejected": -0.09919053316116333,
+      "loss": 0.871,
+      "rewards/accuracies": 0.5416666865348816,
+      "rewards/chosen": -0.21464140713214874,
+      "rewards/margins": -0.01626037061214447,
+      "rewards/rejected": -0.19838106632232666,
+      "step": 53
+    },
+    {
+      "epoch": 0.648,
+      "grad_norm": 0.9795694351196289,
+      "learning_rate": 1.667150513144856e-07,
+      "logits/chosen": -1.3203659057617188,
+      "logits/rejected": -1.1565968990325928,
+      "logps/chosen": -0.09574580192565918,
+      "logps/rejected": -0.09623640775680542,
+      "loss": 0.8547,
+      "rewards/accuracies": 0.5520833730697632,
+      "rewards/chosen": -0.19149160385131836,
+      "rewards/margins": 0.000981215387582779,
+      "rewards/rejected": -0.19247281551361084,
+      "step": 54
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 1.0732730627059937,
+      "learning_rate": 1.5678588055492286e-07,
+      "logits/chosen": -1.2979214191436768,
+      "logits/rejected": -1.150040626525879,
+      "logps/chosen": -0.10067766904830933,
+      "logps/rejected": -0.1018737256526947,
+      "loss": 0.8545,
+      "rewards/accuracies": 0.5104166865348816,
+      "rewards/chosen": -0.20135533809661865,
+      "rewards/margins": 0.0023921187967061996,
+      "rewards/rejected": -0.2037474513053894,
+      "step": 55
+    },
+    {
+      "epoch": 0.672,
+      "grad_norm": 0.9587694406509399,
+      "learning_rate": 1.4702468793900186e-07,
+      "logits/chosen": -1.3606462478637695,
+      "logits/rejected": -1.1573679447174072,
+      "logps/chosen": -0.09973961114883423,
+      "logps/rejected": -0.10131655633449554,
+      "loss": 0.8535,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.19947922229766846,
+      "rewards/margins": 0.003153874073177576,
+      "rewards/rejected": -0.2026331126689911,
+      "step": 56
+    },
+    {
+      "epoch": 0.684,
+      "grad_norm": 1.0398699045181274,
+      "learning_rate": 1.3744906379558164e-07,
+      "logits/chosen": -1.3375906944274902,
+      "logits/rejected": -1.1461068391799927,
+      "logps/chosen": -0.10326816886663437,
+      "logps/rejected": -0.10188017785549164,
+      "loss": 0.857,
+      "rewards/accuracies": 0.4687500298023224,
+      "rewards/chosen": -0.20653633773326874,
+      "rewards/margins": -0.002775975503027439,
+      "rewards/rejected": -0.20376035571098328,
+      "step": 57
+    },
+    {
+      "epoch": 0.696,
+      "grad_norm": 1.0405802726745605,
+      "learning_rate": 1.280762640465914e-07,
+      "logits/chosen": -1.3368816375732422,
+      "logits/rejected": -1.1643409729003906,
+      "logps/chosen": -0.09207028895616531,
+      "logps/rejected": -0.09347332268953323,
+      "loss": 0.8534,
+      "rewards/accuracies": 0.5208333134651184,
+      "rewards/chosen": -0.18414057791233063,
+      "rewards/margins": 0.002806063275784254,
+      "rewards/rejected": -0.18694664537906647,
+      "step": 58
+    },
+    {
+      "epoch": 0.708,
+      "grad_norm": 1.0253372192382812,
+      "learning_rate": 1.189231791106921e-07,
+      "logits/chosen": -1.2978026866912842,
+      "logits/rejected": -1.13652765750885,
+      "logps/chosen": -0.10593652725219727,
+      "logps/rejected": -0.10165742039680481,
+      "loss": 0.8609,
+      "rewards/accuracies": 0.479166716337204,
+      "rewards/chosen": -0.21187305450439453,
+      "rewards/margins": -0.008558189496397972,
+      "rewards/rejected": -0.20331484079360962,
+      "step": 59
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 1.2529534101486206,
+      "learning_rate": 1.1000630346560116e-07,
+      "logits/chosen": -1.3010480403900146,
+      "logits/rejected": -1.133022665977478,
+      "logps/chosen": -0.11126932501792908,
+      "logps/rejected": -0.09786901623010635,
+      "loss": 0.8738,
+      "rewards/accuracies": 0.4895833432674408,
+      "rewards/chosen": -0.22253865003585815,
+      "rewards/margins": -0.026800617575645447,
+      "rewards/rejected": -0.1957380324602127,
+      "step": 60
+    },
+    {
+      "epoch": 0.732,
+      "grad_norm": 0.9133132696151733,
+      "learning_rate": 1.0134170592392836e-07,
+      "logits/chosen": -1.3394014835357666,
+      "logits/rejected": -1.1518943309783936,
+      "logps/chosen": -0.09960196912288666,
+      "logps/rejected": -0.10244160890579224,
+      "loss": 0.8521,
+      "rewards/accuracies": 0.5104166865348816,
+      "rewards/chosen": -0.19920393824577332,
+      "rewards/margins": 0.005679287016391754,
+      "rewards/rejected": -0.20488321781158447,
+      "step": 61
+    },
+    {
+      "epoch": 0.744,
+      "grad_norm": 1.0934334993362427,
+      "learning_rate": 9.29450006760894e-08,
+      "logits/chosen": -1.3294323682785034,
+      "logits/rejected": -1.1431366205215454,
+      "logps/chosen": -0.10188900679349899,
+      "logps/rejected": -0.09653942286968231,
+      "loss": 0.8616,
+      "rewards/accuracies": 0.5,
+      "rewards/chosen": -0.20377801358699799,
+      "rewards/margins": -0.010699168778955936,
+      "rewards/rejected": -0.19307884573936462,
+      "step": 62
+    },
+    {
+      "epoch": 0.756,
+      "grad_norm": 0.9888376593589783,
+      "learning_rate": 8.483131915247967e-08,
+      "logits/chosen": -1.3347070217132568,
+      "logits/rejected": -1.167306661605835,
+      "logps/chosen": -0.10506478697061539,
+      "logps/rejected": -0.1007775291800499,
+      "loss": 0.8603,
+      "rewards/accuracies": 0.5000000596046448,
+      "rewards/chosen": -0.21012957394123077,
+      "rewards/margins": -0.008574524894356728,
+      "rewards/rejected": -0.2015550583600998,
+      "step": 63
+    },
+    {
+      "epoch": 0.768,
+      "grad_norm": 0.9486240744590759,
+      "learning_rate": 7.701528275561347e-08,
+      "logits/chosen": -1.3588067293167114,
+      "logits/rejected": -1.1604478359222412,
+      "logps/chosen": -0.0977708101272583,
+      "logps/rejected": -0.09753303229808807,
+      "loss": 0.8555,
+      "rewards/accuracies": 0.5312500596046448,
+      "rewards/chosen": -0.1955416202545166,
+      "rewards/margins": -0.0004755451809614897,
+      "rewards/rejected": -0.19506606459617615,
+      "step": 64
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 1.033793568611145,
+      "learning_rate": 6.951097651136889e-08,
+      "logits/chosen": -1.3951979875564575,
+      "logits/rejected": -1.125361680984497,
+      "logps/chosen": -0.10307514667510986,
+      "logps/rejected": -0.1052648052573204,
+      "loss": 0.8525,
+      "rewards/accuracies": 0.4895833432674408,
+      "rewards/chosen": -0.20615029335021973,
+      "rewards/margins": 0.0043793064542114735,
+      "rewards/rejected": -0.2105296105146408,
+      "step": 65
+    },
+    {
+      "epoch": 0.792,
+      "grad_norm": 1.0308159589767456,
+      "learning_rate": 6.233192368681889e-08,
+      "logits/chosen": -1.3253390789031982,
+      "logits/rejected": -1.1783477067947388,
+      "logps/chosen": -0.09217476844787598,
+      "logps/rejected": -0.09416632354259491,
+      "loss": 0.8527,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.18434953689575195,
+      "rewards/margins": 0.003983109723776579,
+      "rewards/rejected": -0.18833264708518982,
+      "step": 66
+    },
+    {
+      "epoch": 0.804,
+      "grad_norm": 1.1081663370132446,
+      "learning_rate": 5.5491061420390174e-08,
+      "logits/chosen": -1.2740880250930786,
+      "logits/rejected": -1.152024269104004,
+      "logps/chosen": -0.10923092067241669,
+      "logps/rejected": -0.10176312923431396,
+      "loss": 0.8642,
+      "rewards/accuracies": 0.5104166865348816,
+      "rewards/chosen": -0.21846184134483337,
+      "rewards/margins": -0.01493558008223772,
+      "rewards/rejected": -0.20352625846862793,
+      "step": 67
+    },
+    {
+      "epoch": 0.816,
+      "grad_norm": 1.0150699615478516,
+      "learning_rate": 4.900071740826489e-08,
+      "logits/chosen": -1.348282814025879,
+      "logits/rejected": -1.1383142471313477,
+      "logps/chosen": -0.10064545273780823,
+      "logps/rejected": -0.10860613733530045,
+      "loss": 0.8459,
+      "rewards/accuracies": 0.6041667461395264,
+      "rewards/chosen": -0.20129090547561646,
+      "rewards/margins": 0.01592138595879078,
+      "rewards/rejected": -0.2172122746706009,
+      "step": 68
+    },
+    {
+      "epoch": 0.828,
+      "grad_norm": 1.123404860496521,
+      "learning_rate": 4.287258768903948e-08,
+      "logits/chosen": -1.3442084789276123,
+      "logits/rejected": -1.146917462348938,
+      "logps/chosen": -0.10861781984567642,
+      "logps/rejected": -0.10071661323308945,
+      "loss": 0.8688,
+      "rewards/accuracies": 0.5833333134651184,
+      "rewards/chosen": -0.21723563969135284,
+      "rewards/margins": -0.015802428126335144,
+      "rewards/rejected": -0.2014332264661789,
+      "step": 69
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.9981204867362976,
+      "learning_rate": 3.7117715566672176e-08,
+      "logits/chosen": -1.3524047136306763,
+      "logits/rejected": -1.1452308893203735,
+      "logps/chosen": -0.10508691519498825,
+      "logps/rejected": -0.10403452813625336,
+      "loss": 0.8568,
+      "rewards/accuracies": 0.5729166865348816,
+      "rewards/chosen": -0.2101738303899765,
+      "rewards/margins": -0.0021047808695584536,
+      "rewards/rejected": -0.2080690562725067,
+      "step": 70
+    },
+    {
+      "epoch": 0.852,
+      "grad_norm": 1.0365066528320312,
+      "learning_rate": 3.174647170970296e-08,
+      "logits/chosen": -1.3592528104782104,
+      "logits/rejected": -1.1138660907745361,
+      "logps/chosen": -0.10330555588006973,
+      "logps/rejected": -0.1064281240105629,
+      "loss": 0.8524,
+      "rewards/accuracies": 0.5729166865348816,
+      "rewards/chosen": -0.20661111176013947,
+      "rewards/margins": 0.006245152093470097,
+      "rewards/rejected": -0.2128562480211258,
+      "step": 71
+    },
+    {
+      "epoch": 0.864,
+      "grad_norm": 1.076464295387268,
+      "learning_rate": 2.6768535462607905e-08,
+      "logits/chosen": -1.3058511018753052,
+      "logits/rejected": -1.126082420349121,
+      "logps/chosen": -0.10569944232702255,
+      "logps/rejected": -0.11077728122472763,
+      "loss": 0.8494,
+      "rewards/accuracies": 0.5,
+      "rewards/chosen": -0.2113988846540451,
+      "rewards/margins": 0.01015565823763609,
+      "rewards/rejected": -0.22155456244945526,
+      "step": 72
+    },
+    {
+      "epoch": 0.876,
+      "grad_norm": 0.9264240264892578,
+      "learning_rate": 2.2192877402966048e-08,
+      "logits/chosen": -1.3490333557128906,
+      "logits/rejected": -1.1134393215179443,
+      "logps/chosen": -0.10448520630598068,
+      "logps/rejected": -0.11283887922763824,
+      "loss": 0.8457,
+      "rewards/accuracies": 0.6145833730697632,
+      "rewards/chosen": -0.20897041261196136,
+      "rewards/margins": 0.016707373782992363,
+      "rewards/rejected": -0.2256777584552765,
+      "step": 73
+    },
+    {
+      "epoch": 0.888,
+      "grad_norm": 1.067218542098999,
+      "learning_rate": 1.8027743175872662e-08,
+      "logits/chosen": -1.345249891281128,
+      "logits/rejected": -1.1063634157180786,
+      "logps/chosen": -0.11241614818572998,
+      "logps/rejected": -0.11045221984386444,
+      "loss": 0.8579,
+      "rewards/accuracies": 0.5104166865348816,
+      "rewards/chosen": -0.22483229637145996,
+      "rewards/margins": -0.003927857149392366,
+      "rewards/rejected": -0.22090443968772888,
+      "step": 74
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 1.0710501670837402,
+      "learning_rate": 1.4280638634728948e-08,
+      "logits/chosen": -1.3244132995605469,
+      "logits/rejected": -1.1485953330993652,
+      "logps/chosen": -0.11838357150554657,
+      "logps/rejected": -0.10926854610443115,
+      "loss": 0.8683,
+      "rewards/accuracies": 0.47916674613952637,
+      "rewards/chosen": -0.23676714301109314,
+      "rewards/margins": -0.018230034038424492,
+      "rewards/rejected": -0.2185370922088623,
+      "step": 75
+    },
+    {
+      "epoch": 0.912,
+      "grad_norm": 1.1065670251846313,
+      "learning_rate": 1.0958316315187289e-08,
+      "logits/chosen": -1.317086100578308,
+      "logits/rejected": -1.1413421630859375,
+      "logps/chosen": -0.10257872194051743,
+      "logps/rejected": -0.10372138023376465,
+      "loss": 0.8542,
+      "rewards/accuracies": 0.5416666269302368,
+      "rewards/chosen": -0.20515744388103485,
+      "rewards/margins": 0.0022853193804621696,
+      "rewards/rejected": -0.2074427604675293,
+      "step": 76
+    },
+    {
+      "epoch": 0.924,
+      "grad_norm": 1.0825769901275635,
+      "learning_rate": 8.066763266625282e-09,
+      "logits/chosen": -1.3573386669158936,
+      "logits/rejected": -1.109717607498169,
+      "logps/chosen": -0.10758916288614273,
+      "logps/rejected": -0.10599493980407715,
+      "loss": 0.8597,
+      "rewards/accuracies": 0.6250000596046448,
+      "rewards/chosen": -0.21517832577228546,
+      "rewards/margins": -0.0031884238123893738,
+      "rewards/rejected": -0.2119898796081543,
+      "step": 77
+    },
+    {
+      "epoch": 0.936,
+      "grad_norm": 0.9156083464622498,
+      "learning_rate": 5.611190263078463e-09,
+      "logits/chosen": -1.309991717338562,
+      "logits/rejected": -1.1210401058197021,
+      "logps/chosen": -0.0973024070262909,
+      "logps/rejected": -0.09729278087615967,
+      "loss": 0.8551,
+      "rewards/accuracies": 0.4583333432674408,
+      "rewards/chosen": -0.1946048140525818,
+      "rewards/margins": -1.9263941794633865e-05,
+      "rewards/rejected": -0.19458556175231934,
+      "step": 78
+    },
+    {
+      "epoch": 0.948,
+      "grad_norm": 0.9994163513183594,
+      "learning_rate": 3.5960224130728858e-09,
+      "logits/chosen": -1.3023028373718262,
+      "logits/rejected": -1.1124149560928345,
+      "logps/chosen": -0.09809858351945877,
+      "logps/rejected": -0.1080770194530487,
+      "loss": 0.8436,
+      "rewards/accuracies": 0.6354166865348816,
+      "rewards/chosen": -0.19619716703891754,
+      "rewards/margins": 0.01995689421892166,
+      "rewards/rejected": -0.2161540389060974,
+      "step": 79
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.9938739538192749,
+      "learning_rate": 2.0248911852807917e-09,
+      "logits/chosen": -1.3484057188034058,
+      "logits/rejected": -1.0871906280517578,
+      "logps/chosen": -0.11684219539165497,
+      "logps/rejected": -0.11286689341068268,
+      "loss": 0.8601,
+      "rewards/accuracies": 0.5104166865348816,
+      "rewards/chosen": -0.23368439078330994,
+      "rewards/margins": -0.007950600236654282,
+      "rewards/rejected": -0.22573378682136536,
+      "step": 80
+    },
+    {
+      "epoch": 0.972,
+      "grad_norm": 0.9481876492500305,
+      "learning_rate": 9.006278643683696e-10,
+      "logits/chosen": -1.3169337511062622,
+      "logits/rejected": -1.1376291513442993,
+      "logps/chosen": -0.09936561435461044,
+      "logps/rejected": -0.10312428325414658,
+      "loss": 0.8508,
+      "rewards/accuracies": 0.5833333730697632,
+      "rewards/chosen": -0.1987312287092209,
+      "rewards/margins": 0.007517362013459206,
+      "rewards/rejected": -0.20624856650829315,
+      "step": 81
+    },
+    {
+      "epoch": 0.984,
+      "grad_norm": 1.0620635747909546,
+      "learning_rate": 2.2525844882964606e-10,
+      "logits/chosen": -1.3467659950256348,
+      "logits/rejected": -1.1462361812591553,
+      "logps/chosen": -0.10522940754890442,
+      "logps/rejected": -0.10153805464506149,
+      "loss": 0.8594,
+      "rewards/accuracies": 0.4687500298023224,
+      "rewards/chosen": -0.21045881509780884,
+      "rewards/margins": -0.007382689975202084,
+      "rewards/rejected": -0.20307610929012299,
+      "step": 82
+    },
+    {
+      "epoch": 0.996,
+      "grad_norm": 1.0221400260925293,
+      "learning_rate": 0.0,
+      "logits/chosen": -1.313194751739502,
+      "logits/rejected": -1.164141297340393,
+      "logps/chosen": -0.10121379047632217,
+      "logps/rejected": -0.1018705815076828,
+      "loss": 0.8549,
+      "rewards/accuracies": 0.5416666865348816,
+      "rewards/chosen": -0.20242758095264435,
+      "rewards/margins": 0.0013135506305843592,
+      "rewards/rejected": -0.2037411630153656,
+      "step": 83
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 83,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 305651334512640.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

vocab.json ADDED Viewed

The diff for this file is too large to render. See raw diff