add file

.gitattributes

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

README.md

@@ -0,0 +1,61 @@
+---
+library_name: peft
+license: other
+base_model: /data2/zhuotaotian/psp/llm/utils/models/repo/Qwen2-VL-2B-Instruct
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: Qwen2_VL_2B_Instruct_explore_normal+_6k_removed_2e_6
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# Qwen2_VL_2B_Instruct_explore_normal+_6k_removed_2e_6
+
+This model is a fine-tuned version of [/data2/zhuotaotian/psp/llm/utils/models/repo/Qwen2-VL-2B-Instruct](https://huggingface.co//data2/zhuotaotian/psp/llm/utils/models/repo/Qwen2-VL-2B-Instruct) on the Qwen2_VL_2B_12k dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-06
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 64
+- total_eval_batch_size: 16
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.46.1
+- Pytorch 2.6.0+cu124
+- Datasets 3.1.0
+- Tokenizers 0.20.3

adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/psp/llm/utils/models/repo/Qwen2-VL-2B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "^(?!.*visual.patch_embed|visual.blocks).*(?:down_proj|o_proj|gate_proj|v_proj|q_proj|up_proj|k_proj).*",
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c89a622bd285c45d481d64eb531bec6898253b72740028c67bcd6d55f295b6e2
+size 295489720

added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.9984,
+    "total_flos": 1.3247622626987213e+17,
+    "train_loss": 0.641177637836872,
+    "train_runtime": 1617.7217,
+    "train_samples_per_second": 6.182,
+    "train_steps_per_second": 0.096
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}

special_tokens_map.json

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

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091aa7594dc2fcfbfa06b9e3c22a5f0562ac14f30375c13af7309407a0e67b8a
+size 11420371

tokenizer_config.json

@@ -0,0 +1,144 @@
+{
+  "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
+    }
+  },
+  "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": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 4096,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.9984,
+    "total_flos": 1.3247622626987213e+17,
+    "train_loss": 0.641177637836872,
+    "train_runtime": 1617.7217,
+    "train_samples_per_second": 6.182,
+    "train_steps_per_second": 0.096
+}

trainer_log.jsonl

@@ -0,0 +1,157 @@
+{"current_steps": 1, "total_steps": 156, "loss": 0.6931, "accuracy": 0.0, "lr": 1.9997972289848503e-06, "epoch": 0.0064, "percentage": 0.64, "elapsed_time": "0:00:10", "remaining_time": "0:27:31"}
+{"current_steps": 2, "total_steps": 156, "loss": 0.6929, "accuracy": 0.46875, "lr": 1.9991889981715695e-06, "epoch": 0.0128, "percentage": 1.28, "elapsed_time": "0:00:20", "remaining_time": "0:26:06"}
+{"current_steps": 3, "total_steps": 156, "loss": 0.6937, "accuracy": 0.40625, "lr": 1.9981755542233175e-06, "epoch": 0.0192, "percentage": 1.92, "elapsed_time": "0:00:30", "remaining_time": "0:26:20"}
+{"current_steps": 4, "total_steps": 156, "loss": 0.6928, "accuracy": 0.421875, "lr": 1.99675730813421e-06, "epoch": 0.0256, "percentage": 2.56, "elapsed_time": "0:00:41", "remaining_time": "0:26:12"}
+{"current_steps": 5, "total_steps": 156, "loss": 0.6934, "accuracy": 0.484375, "lr": 1.9949348350626453e-06, "epoch": 0.032, "percentage": 3.21, "elapsed_time": "0:00:51", "remaining_time": "0:26:07"}
+{"current_steps": 6, "total_steps": 156, "loss": 0.6941, "accuracy": 0.390625, "lr": 1.9927088740980536e-06, "epoch": 0.0384, "percentage": 3.85, "elapsed_time": "0:01:02", "remaining_time": "0:25:53"}
+{"current_steps": 7, "total_steps": 156, "loss": 0.6914, "accuracy": 0.546875, "lr": 1.990080327961164e-06, "epoch": 0.0448, "percentage": 4.49, "elapsed_time": "0:01:12", "remaining_time": "0:25:39"}
+{"current_steps": 8, "total_steps": 156, "loss": 0.691, "accuracy": 0.515625, "lr": 1.9870502626379126e-06, "epoch": 0.0512, "percentage": 5.13, "elapsed_time": "0:01:22", "remaining_time": "0:25:32"}
+{"current_steps": 9, "total_steps": 156, "loss": 0.6916, "accuracy": 0.53125, "lr": 1.9836199069471437e-06, "epoch": 0.0576, "percentage": 5.77, "elapsed_time": "0:01:32", "remaining_time": "0:25:14"}
+{"current_steps": 10, "total_steps": 156, "loss": 0.6907, "accuracy": 0.484375, "lr": 1.9797906520422674e-06, "epoch": 0.064, "percentage": 6.41, "elapsed_time": "0:01:43", "remaining_time": "0:25:06"}
+{"current_steps": 11, "total_steps": 156, "loss": 0.6901, "accuracy": 0.5625, "lr": 1.975564050847094e-06, "epoch": 0.0704, "percentage": 7.05, "elapsed_time": "0:01:53", "remaining_time": "0:24:53"}
+{"current_steps": 12, "total_steps": 156, "loss": 0.6861, "accuracy": 0.5625, "lr": 1.970941817426052e-06, "epoch": 0.0768, "percentage": 7.69, "elapsed_time": "0:02:03", "remaining_time": "0:24:38"}
+{"current_steps": 13, "total_steps": 156, "loss": 0.6907, "accuracy": 0.53125, "lr": 1.965925826289068e-06, "epoch": 0.0832, "percentage": 8.33, "elapsed_time": "0:02:13", "remaining_time": "0:24:28"}
+{"current_steps": 14, "total_steps": 156, "loss": 0.6846, "accuracy": 0.625, "lr": 1.9605181116313723e-06, "epoch": 0.0896, "percentage": 8.97, "elapsed_time": "0:02:24", "remaining_time": "0:24:21"}
+{"current_steps": 15, "total_steps": 156, "loss": 0.6893, "accuracy": 0.5, "lr": 1.9547208665085457e-06, "epoch": 0.096, "percentage": 9.62, "elapsed_time": "0:02:34", "remaining_time": "0:24:15"}
+{"current_steps": 16, "total_steps": 156, "loss": 0.69, "accuracy": 0.46875, "lr": 1.9485364419471455e-06, "epoch": 0.1024, "percentage": 10.26, "elapsed_time": "0:02:45", "remaining_time": "0:24:03"}
+{"current_steps": 17, "total_steps": 156, "loss": 0.686, "accuracy": 0.546875, "lr": 1.9419673459912647e-06, "epoch": 0.1088, "percentage": 10.9, "elapsed_time": "0:02:55", "remaining_time": "0:23:55"}
+{"current_steps": 18, "total_steps": 156, "loss": 0.6765, "accuracy": 0.671875, "lr": 1.9350162426854148e-06, "epoch": 0.1152, "percentage": 11.54, "elapsed_time": "0:03:06", "remaining_time": "0:23:46"}
+{"current_steps": 19, "total_steps": 156, "loss": 0.6808, "accuracy": 0.578125, "lr": 1.927685950994143e-06, "epoch": 0.1216, "percentage": 12.18, "elapsed_time": "0:03:16", "remaining_time": "0:23:38"}
+{"current_steps": 20, "total_steps": 156, "loss": 0.6823, "accuracy": 0.640625, "lr": 1.9199794436588242e-06, "epoch": 0.128, "percentage": 12.82, "elapsed_time": "0:03:27", "remaining_time": "0:23:31"}
+{"current_steps": 21, "total_steps": 156, "loss": 0.6839, "accuracy": 0.578125, "lr": 1.91189984599209e-06, "epoch": 0.1344, "percentage": 13.46, "elapsed_time": "0:03:37", "remaining_time": "0:23:19"}
+{"current_steps": 22, "total_steps": 156, "loss": 0.6836, "accuracy": 0.484375, "lr": 1.9034504346103822e-06, "epoch": 0.1408, "percentage": 14.1, "elapsed_time": "0:03:48", "remaining_time": "0:23:09"}
+{"current_steps": 23, "total_steps": 156, "loss": 0.6782, "accuracy": 0.53125, "lr": 1.8946346361051474e-06, "epoch": 0.1472, "percentage": 14.74, "elapsed_time": "0:03:58", "remaining_time": "0:22:56"}
+{"current_steps": 24, "total_steps": 156, "loss": 0.6767, "accuracy": 0.5625, "lr": 1.8854560256532098e-06, "epoch": 0.1536, "percentage": 15.38, "elapsed_time": "0:04:08", "remaining_time": "0:22:45"}
+{"current_steps": 25, "total_steps": 156, "loss": 0.6843, "accuracy": 0.53125, "lr": 1.8759183255668878e-06, "epoch": 0.16, "percentage": 16.03, "elapsed_time": "0:04:19", "remaining_time": "0:22:40"}
+{"current_steps": 26, "total_steps": 156, "loss": 0.6688, "accuracy": 0.671875, "lr": 1.8660254037844386e-06, "epoch": 0.1664, "percentage": 16.67, "elapsed_time": "0:04:30", "remaining_time": "0:22:30"}
+{"current_steps": 27, "total_steps": 156, "loss": 0.6794, "accuracy": 0.59375, "lr": 1.8557812723014475e-06, "epoch": 0.1728, "percentage": 17.31, "elapsed_time": "0:04:40", "remaining_time": "0:22:20"}
+{"current_steps": 28, "total_steps": 156, "loss": 0.6732, "accuracy": 0.59375, "lr": 1.8451900855437948e-06, "epoch": 0.1792, "percentage": 17.95, "elapsed_time": "0:04:51", "remaining_time": "0:22:14"}
+{"current_steps": 29, "total_steps": 156, "loss": 0.6852, "accuracy": 0.53125, "lr": 1.8342561386828613e-06, "epoch": 0.1856, "percentage": 18.59, "elapsed_time": "0:05:02", "remaining_time": "0:22:04"}
+{"current_steps": 30, "total_steps": 156, "loss": 0.6714, "accuracy": 0.625, "lr": 1.8229838658936564e-06, "epoch": 0.192, "percentage": 19.23, "elapsed_time": "0:05:12", "remaining_time": "0:21:53"}
+{"current_steps": 31, "total_steps": 156, "loss": 0.6632, "accuracy": 0.703125, "lr": 1.811377838556573e-06, "epoch": 0.1984, "percentage": 19.87, "elapsed_time": "0:05:23", "remaining_time": "0:21:45"}
+{"current_steps": 32, "total_steps": 156, "loss": 0.6698, "accuracy": 0.578125, "lr": 1.7994427634035012e-06, "epoch": 0.2048, "percentage": 20.51, "elapsed_time": "0:05:34", "remaining_time": "0:21:34"}
+{"current_steps": 33, "total_steps": 156, "loss": 0.6714, "accuracy": 0.625, "lr": 1.78718348060905e-06, "epoch": 0.2112, "percentage": 21.15, "elapsed_time": "0:05:45", "remaining_time": "0:21:27"}
+{"current_steps": 34, "total_steps": 156, "loss": 0.6706, "accuracy": 0.546875, "lr": 1.7746049618276543e-06, "epoch": 0.2176, "percentage": 21.79, "elapsed_time": "0:05:55", "remaining_time": "0:21:16"}
+{"current_steps": 35, "total_steps": 156, "loss": 0.6824, "accuracy": 0.53125, "lr": 1.7617123081773591e-06, "epoch": 0.224, "percentage": 22.44, "elapsed_time": "0:06:07", "remaining_time": "0:21:10"}
+{"current_steps": 36, "total_steps": 156, "loss": 0.6687, "accuracy": 0.546875, "lr": 1.748510748171101e-06, "epoch": 0.2304, "percentage": 23.08, "elapsed_time": "0:06:18", "remaining_time": "0:21:00"}
+{"current_steps": 37, "total_steps": 156, "loss": 0.6701, "accuracy": 0.609375, "lr": 1.7350056355963285e-06, "epoch": 0.2368, "percentage": 23.72, "elapsed_time": "0:06:28", "remaining_time": "0:20:49"}
+{"current_steps": 38, "total_steps": 156, "loss": 0.6709, "accuracy": 0.59375, "lr": 1.7212024473438145e-06, "epoch": 0.2432, "percentage": 24.36, "elapsed_time": "0:06:39", "remaining_time": "0:20:39"}
+{"current_steps": 39, "total_steps": 156, "loss": 0.6688, "accuracy": 0.53125, "lr": 1.7071067811865474e-06, "epoch": 0.2496, "percentage": 25.0, "elapsed_time": "0:06:49", "remaining_time": "0:20:27"}
+{"current_steps": 40, "total_steps": 156, "loss": 0.6685, "accuracy": 0.59375, "lr": 1.6927243535095994e-06, "epoch": 0.256, "percentage": 25.64, "elapsed_time": "0:07:00", "remaining_time": "0:20:19"}
+{"current_steps": 41, "total_steps": 156, "loss": 0.6741, "accuracy": 0.484375, "lr": 1.6780609969918909e-06, "epoch": 0.2624, "percentage": 26.28, "elapsed_time": "0:07:11", "remaining_time": "0:20:10"}
+{"current_steps": 42, "total_steps": 156, "loss": 0.6797, "accuracy": 0.5625, "lr": 1.6631226582407952e-06, "epoch": 0.2688, "percentage": 26.92, "elapsed_time": "0:07:22", "remaining_time": "0:20:00"}
+{"current_steps": 43, "total_steps": 156, "loss": 0.6629, "accuracy": 0.53125, "lr": 1.6479153953805386e-06, "epoch": 0.2752, "percentage": 27.56, "elapsed_time": "0:07:32", "remaining_time": "0:19:49"}
+{"current_steps": 44, "total_steps": 156, "loss": 0.6644, "accuracy": 0.453125, "lr": 1.6324453755953771e-06, "epoch": 0.2816, "percentage": 28.21, "elapsed_time": "0:07:42", "remaining_time": "0:19:37"}
+{"current_steps": 45, "total_steps": 156, "loss": 0.6495, "accuracy": 0.609375, "lr": 1.6167188726285433e-06, "epoch": 0.288, "percentage": 28.85, "elapsed_time": "0:07:52", "remaining_time": "0:19:25"}
+{"current_steps": 46, "total_steps": 156, "loss": 0.6547, "accuracy": 0.640625, "lr": 1.600742264237979e-06, "epoch": 0.2944, "percentage": 29.49, "elapsed_time": "0:08:02", "remaining_time": "0:19:14"}
+{"current_steps": 47, "total_steps": 156, "loss": 0.6512, "accuracy": 0.625, "lr": 1.584522029609889e-06, "epoch": 0.3008, "percentage": 30.13, "elapsed_time": "0:08:11", "remaining_time": "0:19:00"}
+{"current_steps": 48, "total_steps": 156, "loss": 0.6684, "accuracy": 0.546875, "lr": 1.5680647467311555e-06, "epoch": 0.3072, "percentage": 30.77, "elapsed_time": "0:08:21", "remaining_time": "0:18:49"}
+{"current_steps": 49, "total_steps": 156, "loss": 0.6481, "accuracy": 0.65625, "lr": 1.5513770897216919e-06, "epoch": 0.3136, "percentage": 31.41, "elapsed_time": "0:08:32", "remaining_time": "0:18:39"}
+{"current_steps": 50, "total_steps": 156, "loss": 0.6468, "accuracy": 0.65625, "lr": 1.534465826127801e-06, "epoch": 0.32, "percentage": 32.05, "elapsed_time": "0:08:42", "remaining_time": "0:18:27"}
+{"current_steps": 51, "total_steps": 156, "loss": 0.6609, "accuracy": 0.546875, "lr": 1.5173378141776567e-06, "epoch": 0.3264, "percentage": 32.69, "elapsed_time": "0:08:51", "remaining_time": "0:18:14"}
+{"current_steps": 52, "total_steps": 156, "loss": 0.6587, "accuracy": 0.609375, "lr": 1.5e-06, "epoch": 0.3328, "percentage": 33.33, "elapsed_time": "0:09:02", "remaining_time": "0:18:04"}
+{"current_steps": 53, "total_steps": 156, "loss": 0.6477, "accuracy": 0.578125, "lr": 1.4824594148071933e-06, "epoch": 0.3392, "percentage": 33.97, "elapsed_time": "0:09:12", "remaining_time": "0:17:53"}
+{"current_steps": 54, "total_steps": 156, "loss": 0.6565, "accuracy": 0.59375, "lr": 1.4647231720437684e-06, "epoch": 0.3456, "percentage": 34.62, "elapsed_time": "0:09:23", "remaining_time": "0:17:44"}
+{"current_steps": 55, "total_steps": 156, "loss": 0.6516, "accuracy": 0.515625, "lr": 1.4467984645016258e-06, "epoch": 0.352, "percentage": 35.26, "elapsed_time": "0:09:34", "remaining_time": "0:17:34"}
+{"current_steps": 56, "total_steps": 156, "loss": 0.6524, "accuracy": 0.640625, "lr": 1.428692561403054e-06, "epoch": 0.3584, "percentage": 35.9, "elapsed_time": "0:09:44", "remaining_time": "0:17:23"}
+{"current_steps": 57, "total_steps": 156, "loss": 0.6166, "accuracy": 0.671875, "lr": 1.410412805452757e-06, "epoch": 0.3648, "percentage": 36.54, "elapsed_time": "0:09:54", "remaining_time": "0:17:12"}
+{"current_steps": 58, "total_steps": 156, "loss": 0.6464, "accuracy": 0.65625, "lr": 1.391966609860075e-06, "epoch": 0.3712, "percentage": 37.18, "elapsed_time": "0:10:04", "remaining_time": "0:17:01"}
+{"current_steps": 59, "total_steps": 156, "loss": 0.6744, "accuracy": 0.59375, "lr": 1.3733614553326208e-06, "epoch": 0.3776, "percentage": 37.82, "elapsed_time": "0:10:15", "remaining_time": "0:16:51"}
+{"current_steps": 60, "total_steps": 156, "loss": 0.628, "accuracy": 0.6875, "lr": 1.3546048870425354e-06, "epoch": 0.384, "percentage": 38.46, "elapsed_time": "0:10:25", "remaining_time": "0:16:40"}
+{"current_steps": 61, "total_steps": 156, "loss": 0.6312, "accuracy": 0.609375, "lr": 1.3357045115666048e-06, "epoch": 0.3904, "percentage": 39.1, "elapsed_time": "0:10:35", "remaining_time": "0:16:29"}
+{"current_steps": 62, "total_steps": 156, "loss": 0.6324, "accuracy": 0.625, "lr": 1.3166679938014726e-06, "epoch": 0.3968, "percentage": 39.74, "elapsed_time": "0:10:45", "remaining_time": "0:16:18"}
+{"current_steps": 63, "total_steps": 156, "loss": 0.6679, "accuracy": 0.484375, "lr": 1.297503053855203e-06, "epoch": 0.4032, "percentage": 40.38, "elapsed_time": "0:10:55", "remaining_time": "0:16:07"}
+{"current_steps": 64, "total_steps": 156, "loss": 0.6445, "accuracy": 0.609375, "lr": 1.2782174639164526e-06, "epoch": 0.4096, "percentage": 41.03, "elapsed_time": "0:11:05", "remaining_time": "0:15:56"}
+{"current_steps": 65, "total_steps": 156, "loss": 0.6518, "accuracy": 0.625, "lr": 1.2588190451025207e-06, "epoch": 0.416, "percentage": 41.67, "elapsed_time": "0:11:15", "remaining_time": "0:15:45"}
+{"current_steps": 66, "total_steps": 156, "loss": 0.621, "accuracy": 0.6875, "lr": 1.2393156642875577e-06, "epoch": 0.4224, "percentage": 42.31, "elapsed_time": "0:11:25", "remaining_time": "0:15:34"}
+{"current_steps": 67, "total_steps": 156, "loss": 0.6265, "accuracy": 0.640625, "lr": 1.2197152309122171e-06, "epoch": 0.4288, "percentage": 42.95, "elapsed_time": "0:11:35", "remaining_time": "0:15:24"}
+{"current_steps": 68, "total_steps": 156, "loss": 0.6388, "accuracy": 0.609375, "lr": 1.2000256937760445e-06, "epoch": 0.4352, "percentage": 43.59, "elapsed_time": "0:11:45", "remaining_time": "0:15:12"}
+{"current_steps": 69, "total_steps": 156, "loss": 0.6141, "accuracy": 0.65625, "lr": 1.1802550378139057e-06, "epoch": 0.4416, "percentage": 44.23, "elapsed_time": "0:11:55", "remaining_time": "0:15:02"}
+{"current_steps": 70, "total_steps": 156, "loss": 0.6388, "accuracy": 0.546875, "lr": 1.1604112808577602e-06, "epoch": 0.448, "percentage": 44.87, "elapsed_time": "0:12:05", "remaining_time": "0:14:51"}
+{"current_steps": 71, "total_steps": 156, "loss": 0.6052, "accuracy": 0.625, "lr": 1.1405024703850927e-06, "epoch": 0.4544, "percentage": 45.51, "elapsed_time": "0:12:15", "remaining_time": "0:14:40"}
+{"current_steps": 72, "total_steps": 156, "loss": 0.6276, "accuracy": 0.5625, "lr": 1.1205366802553228e-06, "epoch": 0.4608, "percentage": 46.15, "elapsed_time": "0:12:25", "remaining_time": "0:14:29"}
+{"current_steps": 73, "total_steps": 156, "loss": 0.6337, "accuracy": 0.5625, "lr": 1.100522007435517e-06, "epoch": 0.4672, "percentage": 46.79, "elapsed_time": "0:12:34", "remaining_time": "0:14:18"}
+{"current_steps": 74, "total_steps": 156, "loss": 0.6161, "accuracy": 0.640625, "lr": 1.080466568716726e-06, "epoch": 0.4736, "percentage": 47.44, "elapsed_time": "0:12:44", "remaining_time": "0:14:07"}
+{"current_steps": 75, "total_steps": 156, "loss": 0.6425, "accuracy": 0.578125, "lr": 1.060378497422286e-06, "epoch": 0.48, "percentage": 48.08, "elapsed_time": "0:12:55", "remaining_time": "0:13:57"}
+{"current_steps": 76, "total_steps": 156, "loss": 0.6687, "accuracy": 0.5, "lr": 1.0402659401094152e-06, "epoch": 0.4864, "percentage": 48.72, "elapsed_time": "0:13:05", "remaining_time": "0:13:46"}
+{"current_steps": 77, "total_steps": 156, "loss": 0.5872, "accuracy": 0.734375, "lr": 1.0201370532654403e-06, "epoch": 0.4928, "percentage": 49.36, "elapsed_time": "0:13:15", "remaining_time": "0:13:36"}
+{"current_steps": 78, "total_steps": 156, "loss": 0.6213, "accuracy": 0.671875, "lr": 1e-06, "epoch": 0.4992, "percentage": 50.0, "elapsed_time": "0:13:25", "remaining_time": "0:13:25"}
+{"current_steps": 79, "total_steps": 156, "loss": 0.655, "accuracy": 0.578125, "lr": 9.798629467345599e-07, "epoch": 0.5056, "percentage": 50.64, "elapsed_time": "0:13:35", "remaining_time": "0:13:14"}
+{"current_steps": 80, "total_steps": 156, "loss": 0.6179, "accuracy": 0.59375, "lr": 9.597340598905851e-07, "epoch": 0.512, "percentage": 51.28, "elapsed_time": "0:13:45", "remaining_time": "0:13:04"}
+{"current_steps": 81, "total_steps": 156, "loss": 0.6386, "accuracy": 0.515625, "lr": 9.396215025777138e-07, "epoch": 0.5184, "percentage": 51.92, "elapsed_time": "0:13:56", "remaining_time": "0:12:54"}
+{"current_steps": 82, "total_steps": 156, "loss": 0.6184, "accuracy": 0.609375, "lr": 9.195334312832741e-07, "epoch": 0.5248, "percentage": 52.56, "elapsed_time": "0:14:06", "remaining_time": "0:12:44"}
+{"current_steps": 83, "total_steps": 156, "loss": 0.653, "accuracy": 0.5625, "lr": 8.99477992564483e-07, "epoch": 0.5312, "percentage": 53.21, "elapsed_time": "0:14:17", "remaining_time": "0:12:33"}
+{"current_steps": 84, "total_steps": 156, "loss": 0.6248, "accuracy": 0.640625, "lr": 8.79463319744677e-07, "epoch": 0.5376, "percentage": 53.85, "elapsed_time": "0:14:27", "remaining_time": "0:12:23"}
+{"current_steps": 85, "total_steps": 156, "loss": 0.5949, "accuracy": 0.640625, "lr": 8.594975296149075e-07, "epoch": 0.544, "percentage": 54.49, "elapsed_time": "0:14:37", "remaining_time": "0:12:12"}
+{"current_steps": 86, "total_steps": 156, "loss": 0.6498, "accuracy": 0.59375, "lr": 8.395887191422396e-07, "epoch": 0.5504, "percentage": 55.13, "elapsed_time": "0:14:47", "remaining_time": "0:12:02"}
+{"current_steps": 87, "total_steps": 156, "loss": 0.6201, "accuracy": 0.671875, "lr": 8.197449621860942e-07, "epoch": 0.5568, "percentage": 55.77, "elapsed_time": "0:14:57", "remaining_time": "0:11:51"}
+{"current_steps": 88, "total_steps": 156, "loss": 0.6129, "accuracy": 0.5625, "lr": 7.999743062239557e-07, "epoch": 0.5632, "percentage": 56.41, "elapsed_time": "0:15:07", "remaining_time": "0:11:41"}
+{"current_steps": 89, "total_steps": 156, "loss": 0.6035, "accuracy": 0.625, "lr": 7.802847690877831e-07, "epoch": 0.5696, "percentage": 57.05, "elapsed_time": "0:15:17", "remaining_time": "0:11:31"}
+{"current_steps": 90, "total_steps": 156, "loss": 0.6036, "accuracy": 0.640625, "lr": 7.606843357124425e-07, "epoch": 0.576, "percentage": 57.69, "elapsed_time": "0:15:28", "remaining_time": "0:11:20"}
+{"current_steps": 91, "total_steps": 156, "loss": 0.6367, "accuracy": 0.578125, "lr": 7.411809548974791e-07, "epoch": 0.5824, "percentage": 58.33, "elapsed_time": "0:15:38", "remaining_time": "0:11:10"}
+{"current_steps": 92, "total_steps": 156, "loss": 0.5995, "accuracy": 0.625, "lr": 7.217825360835473e-07, "epoch": 0.5888, "percentage": 58.97, "elapsed_time": "0:15:48", "remaining_time": "0:10:59"}
+{"current_steps": 93, "total_steps": 156, "loss": 0.6456, "accuracy": 0.59375, "lr": 7.024969461447972e-07, "epoch": 0.5952, "percentage": 59.62, "elapsed_time": "0:15:59", "remaining_time": "0:10:49"}
+{"current_steps": 94, "total_steps": 156, "loss": 0.6351, "accuracy": 0.5625, "lr": 6.833320061985277e-07, "epoch": 0.6016, "percentage": 60.26, "elapsed_time": "0:16:09", "remaining_time": "0:10:39"}
+{"current_steps": 95, "total_steps": 156, "loss": 0.6033, "accuracy": 0.546875, "lr": 6.642954884333954e-07, "epoch": 0.608, "percentage": 60.9, "elapsed_time": "0:16:21", "remaining_time": "0:10:29"}
+{"current_steps": 96, "total_steps": 156, "loss": 0.6547, "accuracy": 0.515625, "lr": 6.453951129574643e-07, "epoch": 0.6144, "percentage": 61.54, "elapsed_time": "0:16:31", "remaining_time": "0:10:19"}
+{"current_steps": 97, "total_steps": 156, "loss": 0.5912, "accuracy": 0.640625, "lr": 6.26638544667379e-07, "epoch": 0.6208, "percentage": 62.18, "elapsed_time": "0:16:41", "remaining_time": "0:10:09"}
+{"current_steps": 98, "total_steps": 156, "loss": 0.6404, "accuracy": 0.578125, "lr": 6.08033390139925e-07, "epoch": 0.6272, "percentage": 62.82, "elapsed_time": "0:16:52", "remaining_time": "0:09:59"}
+{"current_steps": 99, "total_steps": 156, "loss": 0.6279, "accuracy": 0.515625, "lr": 5.895871945472434e-07, "epoch": 0.6336, "percentage": 63.46, "elapsed_time": "0:17:01", "remaining_time": "0:09:48"}
+{"current_steps": 100, "total_steps": 156, "loss": 0.6504, "accuracy": 0.578125, "lr": 5.713074385969456e-07, "epoch": 0.64, "percentage": 64.1, "elapsed_time": "0:17:11", "remaining_time": "0:09:37"}
+{"current_steps": 101, "total_steps": 156, "loss": 0.6508, "accuracy": 0.609375, "lr": 5.532015354983741e-07, "epoch": 0.6464, "percentage": 64.74, "elapsed_time": "0:17:22", "remaining_time": "0:09:27"}
+{"current_steps": 102, "total_steps": 156, "loss": 0.6362, "accuracy": 0.53125, "lr": 5.352768279562314e-07, "epoch": 0.6528, "percentage": 65.38, "elapsed_time": "0:17:32", "remaining_time": "0:09:17"}
+{"current_steps": 103, "total_steps": 156, "loss": 0.6737, "accuracy": 0.5625, "lr": 5.175405851928068e-07, "epoch": 0.6592, "percentage": 66.03, "elapsed_time": "0:17:42", "remaining_time": "0:09:06"}
+{"current_steps": 104, "total_steps": 156, "loss": 0.578, "accuracy": 0.734375, "lr": 5.000000000000002e-07, "epoch": 0.6656, "percentage": 66.67, "elapsed_time": "0:17:52", "remaining_time": "0:08:56"}
+{"current_steps": 105, "total_steps": 156, "loss": 0.671, "accuracy": 0.515625, "lr": 4.826621858223431e-07, "epoch": 0.672, "percentage": 67.31, "elapsed_time": "0:18:02", "remaining_time": "0:08:45"}
+{"current_steps": 106, "total_steps": 156, "loss": 0.6379, "accuracy": 0.59375, "lr": 4.6553417387219884e-07, "epoch": 0.6784, "percentage": 67.95, "elapsed_time": "0:18:12", "remaining_time": "0:08:35"}
+{"current_steps": 107, "total_steps": 156, "loss": 0.6079, "accuracy": 0.59375, "lr": 4.486229102783083e-07, "epoch": 0.6848, "percentage": 68.59, "elapsed_time": "0:18:23", "remaining_time": "0:08:25"}
+{"current_steps": 108, "total_steps": 156, "loss": 0.6143, "accuracy": 0.625, "lr": 4.3193525326884426e-07, "epoch": 0.6912, "percentage": 69.23, "elapsed_time": "0:18:34", "remaining_time": "0:08:15"}
+{"current_steps": 109, "total_steps": 156, "loss": 0.5908, "accuracy": 0.640625, "lr": 4.1547797039011135e-07, "epoch": 0.6976, "percentage": 69.87, "elapsed_time": "0:18:44", "remaining_time": "0:08:04"}
+{"current_steps": 110, "total_steps": 156, "loss": 0.617, "accuracy": 0.609375, "lr": 3.9925773576202094e-07, "epoch": 0.704, "percentage": 70.51, "elapsed_time": "0:18:55", "remaining_time": "0:07:54"}
+{"current_steps": 111, "total_steps": 156, "loss": 0.6445, "accuracy": 0.5, "lr": 3.832811273714569e-07, "epoch": 0.7104, "percentage": 71.15, "elapsed_time": "0:19:05", "remaining_time": "0:07:44"}
+{"current_steps": 112, "total_steps": 156, "loss": 0.6271, "accuracy": 0.65625, "lr": 3.675546244046228e-07, "epoch": 0.7168, "percentage": 71.79, "elapsed_time": "0:19:16", "remaining_time": "0:07:34"}
+{"current_steps": 113, "total_steps": 156, "loss": 0.6404, "accuracy": 0.578125, "lr": 3.5208460461946133e-07, "epoch": 0.7232, "percentage": 72.44, "elapsed_time": "0:19:26", "remaining_time": "0:07:23"}
+{"current_steps": 114, "total_steps": 156, "loss": 0.6044, "accuracy": 0.59375, "lr": 3.3687734175920503e-07, "epoch": 0.7296, "percentage": 73.08, "elapsed_time": "0:19:37", "remaining_time": "0:07:13"}
+{"current_steps": 115, "total_steps": 156, "loss": 0.5836, "accuracy": 0.671875, "lr": 3.2193900300810904e-07, "epoch": 0.736, "percentage": 73.72, "elapsed_time": "0:19:46", "remaining_time": "0:07:03"}
+{"current_steps": 116, "total_steps": 156, "loss": 0.6137, "accuracy": 0.59375, "lr": 3.072756464904006e-07, "epoch": 0.7424, "percentage": 74.36, "elapsed_time": "0:19:57", "remaining_time": "0:06:52"}
+{"current_steps": 117, "total_steps": 156, "loss": 0.6225, "accuracy": 0.625, "lr": 2.9289321881345254e-07, "epoch": 0.7488, "percentage": 75.0, "elapsed_time": "0:20:08", "remaining_time": "0:06:42"}
+{"current_steps": 118, "total_steps": 156, "loss": 0.597, "accuracy": 0.609375, "lr": 2.7879755265618557e-07, "epoch": 0.7552, "percentage": 75.64, "elapsed_time": "0:20:18", "remaining_time": "0:06:32"}
+{"current_steps": 119, "total_steps": 156, "loss": 0.6394, "accuracy": 0.5625, "lr": 2.649943644036716e-07, "epoch": 0.7616, "percentage": 76.28, "elapsed_time": "0:20:29", "remaining_time": "0:06:22"}
+{"current_steps": 120, "total_steps": 156, "loss": 0.6675, "accuracy": 0.484375, "lr": 2.514892518288988e-07, "epoch": 0.768, "percentage": 76.92, "elapsed_time": "0:20:40", "remaining_time": "0:06:12"}
+{"current_steps": 121, "total_steps": 156, "loss": 0.6162, "accuracy": 0.625, "lr": 2.3828769182264085e-07, "epoch": 0.7744, "percentage": 77.56, "elapsed_time": "0:20:50", "remaining_time": "0:06:01"}
+{"current_steps": 122, "total_steps": 156, "loss": 0.6145, "accuracy": 0.625, "lr": 2.253950381723455e-07, "epoch": 0.7808, "percentage": 78.21, "elapsed_time": "0:21:00", "remaining_time": "0:05:51"}
+{"current_steps": 123, "total_steps": 156, "loss": 0.619, "accuracy": 0.65625, "lr": 2.128165193909499e-07, "epoch": 0.7872, "percentage": 78.85, "elapsed_time": "0:21:10", "remaining_time": "0:05:40"}
+{"current_steps": 124, "total_steps": 156, "loss": 0.6448, "accuracy": 0.546875, "lr": 2.0055723659649905e-07, "epoch": 0.7936, "percentage": 79.49, "elapsed_time": "0:21:20", "remaining_time": "0:05:30"}
+{"current_steps": 125, "total_steps": 156, "loss": 0.6464, "accuracy": 0.625, "lr": 1.8862216144342692e-07, "epoch": 0.8, "percentage": 80.13, "elapsed_time": "0:21:30", "remaining_time": "0:05:20"}
+{"current_steps": 126, "total_steps": 156, "loss": 0.5838, "accuracy": 0.671875, "lr": 1.7701613410634365e-07, "epoch": 0.8064, "percentage": 80.77, "elapsed_time": "0:21:40", "remaining_time": "0:05:09"}
+{"current_steps": 127, "total_steps": 156, "loss": 0.5897, "accuracy": 0.703125, "lr": 1.657438613171387e-07, "epoch": 0.8128, "percentage": 81.41, "elapsed_time": "0:21:51", "remaining_time": "0:04:59"}
+{"current_steps": 128, "total_steps": 156, "loss": 0.6376, "accuracy": 0.515625, "lr": 1.5480991445620538e-07, "epoch": 0.8192, "percentage": 82.05, "elapsed_time": "0:22:01", "remaining_time": "0:04:49"}
+{"current_steps": 129, "total_steps": 156, "loss": 0.6449, "accuracy": 0.640625, "lr": 1.442187276985526e-07, "epoch": 0.8256, "percentage": 82.69, "elapsed_time": "0:22:12", "remaining_time": "0:04:38"}
+{"current_steps": 130, "total_steps": 156, "loss": 0.6029, "accuracy": 0.59375, "lr": 1.3397459621556128e-07, "epoch": 0.832, "percentage": 83.33, "elapsed_time": "0:22:24", "remaining_time": "0:04:28"}
+{"current_steps": 131, "total_steps": 156, "loss": 0.6666, "accuracy": 0.5625, "lr": 1.2408167443311213e-07, "epoch": 0.8384, "percentage": 83.97, "elapsed_time": "0:22:34", "remaining_time": "0:04:18"}
+{"current_steps": 132, "total_steps": 156, "loss": 0.6471, "accuracy": 0.546875, "lr": 1.145439743467902e-07, "epoch": 0.8448, "percentage": 84.62, "elapsed_time": "0:22:45", "remaining_time": "0:04:08"}
+{"current_steps": 133, "total_steps": 156, "loss": 0.6142, "accuracy": 0.59375, "lr": 1.0536536389485273e-07, "epoch": 0.8512, "percentage": 85.26, "elapsed_time": "0:22:55", "remaining_time": "0:03:57"}
+{"current_steps": 134, "total_steps": 156, "loss": 0.589, "accuracy": 0.671875, "lr": 9.65495653896179e-08, "epoch": 0.8576, "percentage": 85.9, "elapsed_time": "0:23:06", "remaining_time": "0:03:47"}
+{"current_steps": 135, "total_steps": 156, "loss": 0.5606, "accuracy": 0.703125, "lr": 8.810015400790993e-08, "epoch": 0.864, "percentage": 86.54, "elapsed_time": "0:23:16", "remaining_time": "0:03:37"}
+{"current_steps": 136, "total_steps": 156, "loss": 0.5921, "accuracy": 0.609375, "lr": 8.002055634117577e-08, "epoch": 0.8704, "percentage": 87.18, "elapsed_time": "0:23:26", "remaining_time": "0:03:26"}
+{"current_steps": 137, "total_steps": 156, "loss": 0.5875, "accuracy": 0.734375, "lr": 7.231404900585714e-08, "epoch": 0.8768, "percentage": 87.82, "elapsed_time": "0:23:36", "remaining_time": "0:03:16"}
+{"current_steps": 138, "total_steps": 156, "loss": 0.5893, "accuracy": 0.671875, "lr": 6.498375731458527e-08, "epoch": 0.8832, "percentage": 88.46, "elapsed_time": "0:23:47", "remaining_time": "0:03:06"}
+{"current_steps": 139, "total_steps": 156, "loss": 0.6445, "accuracy": 0.578125, "lr": 5.8032654008735136e-08, "epoch": 0.8896, "percentage": 89.1, "elapsed_time": "0:23:57", "remaining_time": "0:02:55"}
+{"current_steps": 140, "total_steps": 156, "loss": 0.5961, "accuracy": 0.578125, "lr": 5.146355805285451e-08, "epoch": 0.896, "percentage": 89.74, "elapsed_time": "0:24:07", "remaining_time": "0:02:45"}
+{"current_steps": 141, "total_steps": 156, "loss": 0.5806, "accuracy": 0.65625, "lr": 4.52791334914544e-08, "epoch": 0.9024, "percentage": 90.38, "elapsed_time": "0:24:18", "remaining_time": "0:02:35"}
+{"current_steps": 142, "total_steps": 156, "loss": 0.5824, "accuracy": 0.703125, "lr": 3.948188836862776e-08, "epoch": 0.9088, "percentage": 91.03, "elapsed_time": "0:24:28", "remaining_time": "0:02:24"}
+{"current_steps": 143, "total_steps": 156, "loss": 0.6123, "accuracy": 0.59375, "lr": 3.4074173710931796e-08, "epoch": 0.9152, "percentage": 91.67, "elapsed_time": "0:24:39", "remaining_time": "0:02:14"}
+{"current_steps": 144, "total_steps": 156, "loss": 0.6129, "accuracy": 0.609375, "lr": 2.9058182573947986e-08, "epoch": 0.9216, "percentage": 92.31, "elapsed_time": "0:24:48", "remaining_time": "0:02:04"}
+{"current_steps": 145, "total_steps": 156, "loss": 0.5799, "accuracy": 0.59375, "lr": 2.4435949152906144e-08, "epoch": 0.928, "percentage": 92.95, "elapsed_time": "0:24:59", "remaining_time": "0:01:53"}
+{"current_steps": 146, "total_steps": 156, "loss": 0.6297, "accuracy": 0.515625, "lr": 2.0209347957732326e-08, "epoch": 0.9344, "percentage": 93.59, "elapsed_time": "0:25:09", "remaining_time": "0:01:43"}
+{"current_steps": 147, "total_steps": 156, "loss": 0.5892, "accuracy": 0.625, "lr": 1.6380093052856482e-08, "epoch": 0.9408, "percentage": 94.23, "elapsed_time": "0:25:20", "remaining_time": "0:01:33"}
+{"current_steps": 148, "total_steps": 156, "loss": 0.5887, "accuracy": 0.65625, "lr": 1.2949737362087154e-08, "epoch": 0.9472, "percentage": 94.87, "elapsed_time": "0:25:30", "remaining_time": "0:01:22"}
+{"current_steps": 149, "total_steps": 156, "loss": 0.5811, "accuracy": 0.65625, "lr": 9.919672038835925e-09, "epoch": 0.9536, "percentage": 95.51, "elapsed_time": "0:25:41", "remaining_time": "0:01:12"}
+{"current_steps": 150, "total_steps": 156, "loss": 0.6039, "accuracy": 0.640625, "lr": 7.291125901946027e-09, "epoch": 0.96, "percentage": 96.15, "elapsed_time": "0:25:51", "remaining_time": "0:01:02"}
+{"current_steps": 151, "total_steps": 156, "loss": 0.6305, "accuracy": 0.546875, "lr": 5.065164937354427e-09, "epoch": 0.9664, "percentage": 96.79, "elapsed_time": "0:26:01", "remaining_time": "0:00:51"}
+{"current_steps": 152, "total_steps": 156, "loss": 0.5911, "accuracy": 0.65625, "lr": 3.2426918657900704e-09, "epoch": 0.9728, "percentage": 97.44, "elapsed_time": "0:26:10", "remaining_time": "0:00:41"}
+{"current_steps": 153, "total_steps": 156, "loss": 0.6273, "accuracy": 0.609375, "lr": 1.824445776682504e-09, "epoch": 0.9792, "percentage": 98.08, "elapsed_time": "0:26:20", "remaining_time": "0:00:30"}
+{"current_steps": 154, "total_steps": 156, "loss": 0.5845, "accuracy": 0.625, "lr": 8.110018284304132e-10, "epoch": 0.9856, "percentage": 98.72, "elapsed_time": "0:26:30", "remaining_time": "0:00:20"}
+{"current_steps": 155, "total_steps": 156, "loss": 0.6496, "accuracy": 0.578125, "lr": 2.0277101514987183e-10, "epoch": 0.992, "percentage": 99.36, "elapsed_time": "0:26:41", "remaining_time": "0:00:10"}
+{"current_steps": 156, "total_steps": 156, "loss": 0.6656, "accuracy": 0.5625, "lr": 0.0, "epoch": 0.9984, "percentage": 100.0, "elapsed_time": "0:26:52", "remaining_time": "0:00:00"}
+{"current_steps": 156, "total_steps": 156, "epoch": 0.9984, "percentage": 100.0, "elapsed_time": "0:26:57", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,2382 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9984,
+  "eval_steps": 500,
+  "global_step": 156,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0064,
+      "grad_norm": 1.5359957218170166,
+      "learning_rate": 1.9997972289848503e-06,
+      "logits/chosen": -2.5997238159179688,
+      "logits/rejected": -2.6029367446899414,
+      "logps/chosen": -17.947940826416016,
+      "logps/rejected": -21.561464309692383,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.0128,
+      "grad_norm": 1.6351780891418457,
+      "learning_rate": 1.9991889981715695e-06,
+      "logits/chosen": -2.588522434234619,
+      "logits/rejected": -2.5925989151000977,
+      "logps/chosen": -24.20566177368164,
+      "logps/rejected": -28.453519821166992,
+      "loss": 0.6929,
+      "rewards/accuracies": 0.46875,
+      "rewards/chosen": -0.0015883638989180326,
+      "rewards/margins": 0.0005781143554486334,
+      "rewards/rejected": -0.002166478196159005,
+      "step": 2
+    },
+    {
+      "epoch": 0.0192,
+      "grad_norm": 1.6481831073760986,
+      "learning_rate": 1.9981755542233175e-06,
+      "logits/chosen": -2.5486087799072266,
+      "logits/rejected": -2.553159713745117,
+      "logps/chosen": -21.488862991333008,
+      "logps/rejected": -25.426105499267578,
+      "loss": 0.6937,
+      "rewards/accuracies": 0.40625,
+      "rewards/chosen": -0.0020689992234110832,
+      "rewards/margins": -0.0010093210730701685,
+      "rewards/rejected": -0.0010596781503409147,
+      "step": 3
+    },
+    {
+      "epoch": 0.0256,
+      "grad_norm": 1.85158371925354,
+      "learning_rate": 1.99675730813421e-06,
+      "logits/chosen": -2.5916860103607178,
+      "logits/rejected": -2.5861847400665283,
+      "logps/chosen": -23.558454513549805,
+      "logps/rejected": -26.246479034423828,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.421875,
+      "rewards/chosen": -0.0018257580231875181,
+      "rewards/margins": 0.0008337950566783547,
+      "rewards/rejected": -0.0026595531962811947,
+      "step": 4
+    },
+    {
+      "epoch": 0.032,
+      "grad_norm": 1.7252857685089111,
+      "learning_rate": 1.9949348350626453e-06,
+      "logits/chosen": -2.626497507095337,
+      "logits/rejected": -2.6224427223205566,
+      "logps/chosen": -21.36603546142578,
+      "logps/rejected": -25.15032958984375,
+      "loss": 0.6934,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.004167431965470314,
+      "rewards/margins": -0.00041358935413882136,
+      "rewards/rejected": -0.0037538427859544754,
+      "step": 5
+    },
+    {
+      "epoch": 0.0384,
+      "grad_norm": 1.769866704940796,
+      "learning_rate": 1.9927088740980536e-06,
+      "logits/chosen": -2.6353015899658203,
+      "logits/rejected": -2.6331610679626465,
+      "logps/chosen": -26.10724639892578,
+      "logps/rejected": -31.444061279296875,
+      "loss": 0.6941,
+      "rewards/accuracies": 0.390625,
+      "rewards/chosen": -0.005395121872425079,
+      "rewards/margins": -0.0017277344595640898,
+      "rewards/rejected": -0.0036673876456916332,
+      "step": 6
+    },
+    {
+      "epoch": 0.0448,
+      "grad_norm": 1.6636605262756348,
+      "learning_rate": 1.990080327961164e-06,
+      "logits/chosen": -2.5836739540100098,
+      "logits/rejected": -2.5827889442443848,
+      "logps/chosen": -25.18962287902832,
+      "logps/rejected": -27.809450149536133,
+      "loss": 0.6914,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.0016480791382491589,
+      "rewards/margins": 0.003761656116694212,
+      "rewards/rejected": -0.005409735720604658,
+      "step": 7
+    },
+    {
+      "epoch": 0.0512,
+      "grad_norm": 1.5588624477386475,
+      "learning_rate": 1.9870502626379126e-06,
+      "logits/chosen": -2.5598201751708984,
+      "logits/rejected": -2.559025764465332,
+      "logps/chosen": -21.595394134521484,
+      "logps/rejected": -25.79460906982422,
+      "loss": 0.691,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.005604313686490059,
+      "rewards/margins": 0.00436945678666234,
+      "rewards/rejected": -0.009973770007491112,
+      "step": 8
+    },
+    {
+      "epoch": 0.0576,
+      "grad_norm": 1.6384012699127197,
+      "learning_rate": 1.9836199069471437e-06,
+      "logits/chosen": -2.587116003036499,
+      "logits/rejected": -2.595010757446289,
+      "logps/chosen": -32.334144592285156,
+      "logps/rejected": -32.41010665893555,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.01219846773892641,
+      "rewards/margins": 0.0032913186587393284,
+      "rewards/rejected": -0.015489785000681877,
+      "step": 9
+    },
+    {
+      "epoch": 0.064,
+      "grad_norm": 1.5940217971801758,
+      "learning_rate": 1.9797906520422674e-06,
+      "logits/chosen": -2.5751521587371826,
+      "logits/rejected": -2.577556848526001,
+      "logps/chosen": -25.363061904907227,
+      "logps/rejected": -30.96901512145996,
+      "loss": 0.6907,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.00955671165138483,
+      "rewards/margins": 0.004959252662956715,
+      "rewards/rejected": -0.014515964314341545,
+      "step": 10
+    },
+    {
+      "epoch": 0.0704,
+      "grad_norm": 1.541799545288086,
+      "learning_rate": 1.975564050847094e-06,
+      "logits/chosen": -2.590721368789673,
+      "logits/rejected": -2.591123104095459,
+      "logps/chosen": -32.15184783935547,
+      "logps/rejected": -33.79652404785156,
+      "loss": 0.6901,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.011551897041499615,
+      "rewards/margins": 0.0061759669333696365,
+      "rewards/rejected": -0.017727863043546677,
+      "step": 11
+    },
+    {
+      "epoch": 0.0768,
+      "grad_norm": 1.8116754293441772,
+      "learning_rate": 1.970941817426052e-06,
+      "logits/chosen": -2.5945043563842773,
+      "logits/rejected": -2.59816837310791,
+      "logps/chosen": -28.759960174560547,
+      "logps/rejected": -38.848846435546875,
+      "loss": 0.6861,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.010900590568780899,
+      "rewards/margins": 0.01439574733376503,
+      "rewards/rejected": -0.02529633790254593,
+      "step": 12
+    },
+    {
+      "epoch": 0.0832,
+      "grad_norm": 1.5430954694747925,
+      "learning_rate": 1.965925826289068e-06,
+      "logits/chosen": -2.6023316383361816,
+      "logits/rejected": -2.609272003173828,
+      "logps/chosen": -19.806459426879883,
+      "logps/rejected": -22.50006866455078,
+      "loss": 0.6907,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.010169017128646374,
+      "rewards/margins": 0.005155025515705347,
+      "rewards/rejected": -0.015324043110013008,
+      "step": 13
+    },
+    {
+      "epoch": 0.0896,
+      "grad_norm": 1.7522083520889282,
+      "learning_rate": 1.9605181116313723e-06,
+      "logits/chosen": -2.655667304992676,
+      "logits/rejected": -2.6485490798950195,
+      "logps/chosen": -29.75261688232422,
+      "logps/rejected": -30.21455955505371,
+      "loss": 0.6846,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.01066405326128006,
+      "rewards/margins": 0.017498096451163292,
+      "rewards/rejected": -0.0281621515750885,
+      "step": 14
+    },
+    {
+      "epoch": 0.096,
+      "grad_norm": 1.7028992176055908,
+      "learning_rate": 1.9547208665085457e-06,
+      "logits/chosen": -2.553605556488037,
+      "logits/rejected": -2.5524120330810547,
+      "logps/chosen": -25.244916915893555,
+      "logps/rejected": -24.222930908203125,
+      "loss": 0.6893,
+      "rewards/accuracies": 0.5,
+      "rewards/chosen": -0.018998706713318825,
+      "rewards/margins": 0.008375532925128937,
+      "rewards/rejected": -0.027374237775802612,
+      "step": 15
+    },
+    {
+      "epoch": 0.1024,
+      "grad_norm": 1.4704949855804443,
+      "learning_rate": 1.9485364419471455e-06,
+      "logits/chosen": -2.7163968086242676,
+      "logits/rejected": -2.711620569229126,
+      "logps/chosen": -24.197038650512695,
+      "logps/rejected": -23.20355796813965,
+      "loss": 0.69,
+      "rewards/accuracies": 0.46875,
+      "rewards/chosen": -0.014609398320317268,
+      "rewards/margins": 0.006537495646625757,
+      "rewards/rejected": -0.02114689350128174,
+      "step": 16
+    },
+    {
+      "epoch": 0.1088,
+      "grad_norm": 1.5888653993606567,
+      "learning_rate": 1.9419673459912647e-06,
+      "logits/chosen": -2.565908193588257,
+      "logits/rejected": -2.568258762359619,
+      "logps/chosen": -23.84603500366211,
+      "logps/rejected": -27.283565521240234,
+      "loss": 0.686,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.01647404208779335,
+      "rewards/margins": 0.014618257991969585,
+      "rewards/rejected": -0.03109230101108551,
+      "step": 17
+    },
+    {
+      "epoch": 0.1152,
+      "grad_norm": 1.9215878248214722,
+      "learning_rate": 1.9350162426854148e-06,
+      "logits/chosen": -2.605273962020874,
+      "logits/rejected": -2.600682497024536,
+      "logps/chosen": -24.307863235473633,
+      "logps/rejected": -37.23108673095703,
+      "loss": 0.6765,
+      "rewards/accuracies": 0.671875,
+      "rewards/chosen": -0.02101895771920681,
+      "rewards/margins": 0.03534526377916336,
+      "rewards/rejected": -0.05636422336101532,
+      "step": 18
+    },
+    {
+      "epoch": 0.1216,
+      "grad_norm": 1.5945792198181152,
+      "learning_rate": 1.927685950994143e-06,
+      "logits/chosen": -2.5861587524414062,
+      "logits/rejected": -2.590471029281616,
+      "logps/chosen": -24.360841751098633,
+      "logps/rejected": -29.96798324584961,
+      "loss": 0.6808,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.017002003267407417,
+      "rewards/margins": 0.025491666048765182,
+      "rewards/rejected": -0.04249367117881775,
+      "step": 19
+    },
+    {
+      "epoch": 0.128,
+      "grad_norm": 1.5446585416793823,
+      "learning_rate": 1.9199794436588242e-06,
+      "logits/chosen": -2.590156078338623,
+      "logits/rejected": -2.591686487197876,
+      "logps/chosen": -18.158397674560547,
+      "logps/rejected": -27.70963478088379,
+      "loss": 0.6823,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.019463814795017242,
+      "rewards/margins": 0.022219937294721603,
+      "rewards/rejected": -0.041683752089738846,
+      "step": 20
+    },
+    {
+      "epoch": 0.1344,
+      "grad_norm": 1.650801181793213,
+      "learning_rate": 1.91189984599209e-06,
+      "logits/chosen": -2.6143407821655273,
+      "logits/rejected": -2.616335868835449,
+      "logps/chosen": -28.143112182617188,
+      "logps/rejected": -34.50914764404297,
+      "loss": 0.6839,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.03160880133509636,
+      "rewards/margins": 0.019251443445682526,
+      "rewards/rejected": -0.050860244780778885,
+      "step": 21
+    },
+    {
+      "epoch": 0.1408,
+      "grad_norm": 1.5145279169082642,
+      "learning_rate": 1.9034504346103822e-06,
+      "logits/chosen": -2.5937106609344482,
+      "logits/rejected": -2.5988285541534424,
+      "logps/chosen": -18.56326675415039,
+      "logps/rejected": -24.249303817749023,
+      "loss": 0.6836,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.022663850337266922,
+      "rewards/margins": 0.02033817768096924,
+      "rewards/rejected": -0.04300203174352646,
+      "step": 22
+    },
+    {
+      "epoch": 0.1472,
+      "grad_norm": 1.5345410108566284,
+      "learning_rate": 1.8946346361051474e-06,
+      "logits/chosen": -2.615969657897949,
+      "logits/rejected": -2.621823310852051,
+      "logps/chosen": -29.635540008544922,
+      "logps/rejected": -31.84821319580078,
+      "loss": 0.6782,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.03998611494898796,
+      "rewards/margins": 0.03097831830382347,
+      "rewards/rejected": -0.07096443325281143,
+      "step": 23
+    },
+    {
+      "epoch": 0.1536,
+      "grad_norm": 1.7157692909240723,
+      "learning_rate": 1.8854560256532098e-06,
+      "logits/chosen": -2.620194911956787,
+      "logits/rejected": -2.620255947113037,
+      "logps/chosen": -17.377857208251953,
+      "logps/rejected": -23.9888973236084,
+      "loss": 0.6767,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.020580505952239037,
+      "rewards/margins": 0.034580424427986145,
+      "rewards/rejected": -0.05516092851758003,
+      "step": 24
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 1.6869728565216064,
+      "learning_rate": 1.8759183255668878e-06,
+      "logits/chosen": -2.611105442047119,
+      "logits/rejected": -2.61867618560791,
+      "logps/chosen": -20.499692916870117,
+      "logps/rejected": -24.374731063842773,
+      "loss": 0.6843,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.036568447947502136,
+      "rewards/margins": 0.01921386644244194,
+      "rewards/rejected": -0.055782318115234375,
+      "step": 25
+    },
+    {
+      "epoch": 0.1664,
+      "grad_norm": 1.7368873357772827,
+      "learning_rate": 1.8660254037844386e-06,
+      "logits/chosen": -2.6423802375793457,
+      "logits/rejected": -2.6456449031829834,
+      "logps/chosen": -21.891403198242188,
+      "logps/rejected": -29.566143035888672,
+      "loss": 0.6688,
+      "rewards/accuracies": 0.671875,
+      "rewards/chosen": -0.031646013259887695,
+      "rewards/margins": 0.05181672424077988,
+      "rewards/rejected": -0.08346273750066757,
+      "step": 26
+    },
+    {
+      "epoch": 0.1728,
+      "grad_norm": 1.5447808504104614,
+      "learning_rate": 1.8557812723014475e-06,
+      "logits/chosen": -2.6391639709472656,
+      "logits/rejected": -2.6352741718292236,
+      "logps/chosen": -21.196420669555664,
+      "logps/rejected": -23.61417579650879,
+      "loss": 0.6794,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.03798701614141464,
+      "rewards/margins": 0.02939746156334877,
+      "rewards/rejected": -0.06738448143005371,
+      "step": 27
+    },
+    {
+      "epoch": 0.1792,
+      "grad_norm": 1.6050820350646973,
+      "learning_rate": 1.8451900855437948e-06,
+      "logits/chosen": -2.5705482959747314,
+      "logits/rejected": -2.576892852783203,
+      "logps/chosen": -16.8795166015625,
+      "logps/rejected": -26.7379150390625,
+      "loss": 0.6732,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.029454590752720833,
+      "rewards/margins": 0.04180760681629181,
+      "rewards/rejected": -0.07126219570636749,
+      "step": 28
+    },
+    {
+      "epoch": 0.1856,
+      "grad_norm": 1.6299959421157837,
+      "learning_rate": 1.8342561386828613e-06,
+      "logits/chosen": -2.5929720401763916,
+      "logits/rejected": -2.5860238075256348,
+      "logps/chosen": -31.573650360107422,
+      "logps/rejected": -28.090660095214844,
+      "loss": 0.6852,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.07019858062267303,
+      "rewards/margins": 0.018037281930446625,
+      "rewards/rejected": -0.08823586255311966,
+      "step": 29
+    },
+    {
+      "epoch": 0.192,
+      "grad_norm": 1.6733273267745972,
+      "learning_rate": 1.8229838658936564e-06,
+      "logits/chosen": -2.662106513977051,
+      "logits/rejected": -2.661245822906494,
+      "logps/chosen": -22.231027603149414,
+      "logps/rejected": -27.283416748046875,
+      "loss": 0.6714,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.04910629242658615,
+      "rewards/margins": 0.047111593186855316,
+      "rewards/rejected": -0.09621788561344147,
+      "step": 30
+    },
+    {
+      "epoch": 0.1984,
+      "grad_norm": 1.764510989189148,
+      "learning_rate": 1.811377838556573e-06,
+      "logits/chosen": -2.6185505390167236,
+      "logits/rejected": -2.6172585487365723,
+      "logps/chosen": -21.329898834228516,
+      "logps/rejected": -28.52274513244629,
+      "loss": 0.6632,
+      "rewards/accuracies": 0.703125,
+      "rewards/chosen": -0.051290154457092285,
+      "rewards/margins": 0.06304699182510376,
+      "rewards/rejected": -0.11433714628219604,
+      "step": 31
+    },
+    {
+      "epoch": 0.2048,
+      "grad_norm": 1.6998153924942017,
+      "learning_rate": 1.7994427634035012e-06,
+      "logits/chosen": -2.6329164505004883,
+      "logits/rejected": -2.634524345397949,
+      "logps/chosen": -22.047428131103516,
+      "logps/rejected": -28.356555938720703,
+      "loss": 0.6698,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.045766860246658325,
+      "rewards/margins": 0.051489029079675674,
+      "rewards/rejected": -0.0972558856010437,
+      "step": 32
+    },
+    {
+      "epoch": 0.2112,
+      "grad_norm": 1.5328950881958008,
+      "learning_rate": 1.78718348060905e-06,
+      "logits/chosen": -2.580425262451172,
+      "logits/rejected": -2.5789906978607178,
+      "logps/chosen": -26.442665100097656,
+      "logps/rejected": -32.58696746826172,
+      "loss": 0.6714,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.0714339092373848,
+      "rewards/margins": 0.048205502331256866,
+      "rewards/rejected": -0.11963941156864166,
+      "step": 33
+    },
+    {
+      "epoch": 0.2176,
+      "grad_norm": 1.5043951272964478,
+      "learning_rate": 1.7746049618276543e-06,
+      "logits/chosen": -2.6766915321350098,
+      "logits/rejected": -2.6742377281188965,
+      "logps/chosen": -20.243959426879883,
+      "logps/rejected": -25.328256607055664,
+      "loss": 0.6706,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.0511956512928009,
+      "rewards/margins": 0.04972216486930847,
+      "rewards/rejected": -0.10091781616210938,
+      "step": 34
+    },
+    {
+      "epoch": 0.224,
+      "grad_norm": 1.4899978637695312,
+      "learning_rate": 1.7617123081773591e-06,
+      "logits/chosen": -2.6175575256347656,
+      "logits/rejected": -2.6170494556427,
+      "logps/chosen": -21.302318572998047,
+      "logps/rejected": -22.476587295532227,
+      "loss": 0.6824,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.0734604224562645,
+      "rewards/margins": 0.026149626821279526,
+      "rewards/rejected": -0.09961004555225372,
+      "step": 35
+    },
+    {
+      "epoch": 0.2304,
+      "grad_norm": 1.628516674041748,
+      "learning_rate": 1.748510748171101e-06,
+      "logits/chosen": -2.5839972496032715,
+      "logits/rejected": -2.5774171352386475,
+      "logps/chosen": -28.922618865966797,
+      "logps/rejected": -30.27091407775879,
+      "loss": 0.6687,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.07131488621234894,
+      "rewards/margins": 0.0522158220410347,
+      "rewards/rejected": -0.12353070080280304,
+      "step": 36
+    },
+    {
+      "epoch": 0.2368,
+      "grad_norm": 1.8524824380874634,
+      "learning_rate": 1.7350056355963285e-06,
+      "logits/chosen": -2.614508628845215,
+      "logits/rejected": -2.61613392829895,
+      "logps/chosen": -30.552379608154297,
+      "logps/rejected": -33.45857238769531,
+      "loss": 0.6701,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.08760587871074677,
+      "rewards/margins": 0.055143099278211594,
+      "rewards/rejected": -0.14274898171424866,
+      "step": 37
+    },
+    {
+      "epoch": 0.2432,
+      "grad_norm": 1.5397086143493652,
+      "learning_rate": 1.7212024473438145e-06,
+      "logits/chosen": -2.650211811065674,
+      "logits/rejected": -2.655029296875,
+      "logps/chosen": -21.308271408081055,
+      "logps/rejected": -29.745555877685547,
+      "loss": 0.6709,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.0855458676815033,
+      "rewards/margins": 0.04927408695220947,
+      "rewards/rejected": -0.13481995463371277,
+      "step": 38
+    },
+    {
+      "epoch": 0.2496,
+      "grad_norm": 1.5311026573181152,
+      "learning_rate": 1.7071067811865474e-06,
+      "logits/chosen": -2.675621509552002,
+      "logits/rejected": -2.673921585083008,
+      "logps/chosen": -29.99221420288086,
+      "logps/rejected": -35.94572830200195,
+      "loss": 0.6688,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.12128478288650513,
+      "rewards/margins": 0.054806165397167206,
+      "rewards/rejected": -0.17609095573425293,
+      "step": 39
+    },
+    {
+      "epoch": 0.256,
+      "grad_norm": 1.447879672050476,
+      "learning_rate": 1.6927243535095994e-06,
+      "logits/chosen": -2.60408353805542,
+      "logits/rejected": -2.6094746589660645,
+      "logps/chosen": -18.237171173095703,
+      "logps/rejected": -22.917400360107422,
+      "loss": 0.6685,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.06733156740665436,
+      "rewards/margins": 0.054190389811992645,
+      "rewards/rejected": -0.121521957218647,
+      "step": 40
+    },
+    {
+      "epoch": 0.2624,
+      "grad_norm": 1.4824601411819458,
+      "learning_rate": 1.6780609969918909e-06,
+      "logits/chosen": -2.627382755279541,
+      "logits/rejected": -2.627575159072876,
+      "logps/chosen": -29.326995849609375,
+      "logps/rejected": -31.696285247802734,
+      "loss": 0.6741,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.10711199045181274,
+      "rewards/margins": 0.04250992089509964,
+      "rewards/rejected": -0.1496219038963318,
+      "step": 41
+    },
+    {
+      "epoch": 0.2688,
+      "grad_norm": 1.870728611946106,
+      "learning_rate": 1.6631226582407952e-06,
+      "logits/chosen": -2.6964638233184814,
+      "logits/rejected": -2.6928329467773438,
+      "logps/chosen": -26.665050506591797,
+      "logps/rejected": -20.22939682006836,
+      "loss": 0.6797,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.08645758032798767,
+      "rewards/margins": 0.034841664135456085,
+      "rewards/rejected": -0.12129924446344376,
+      "step": 42
+    },
+    {
+      "epoch": 0.2752,
+      "grad_norm": 1.4833208322525024,
+      "learning_rate": 1.6479153953805386e-06,
+      "logits/chosen": -2.6245405673980713,
+      "logits/rejected": -2.62449049949646,
+      "logps/chosen": -18.757720947265625,
+      "logps/rejected": -24.58340072631836,
+      "loss": 0.6629,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.056469254195690155,
+      "rewards/margins": 0.0671435073018074,
+      "rewards/rejected": -0.12361276149749756,
+      "step": 43
+    },
+    {
+      "epoch": 0.2816,
+      "grad_norm": 1.4494824409484863,
+      "learning_rate": 1.6324453755953771e-06,
+      "logits/chosen": -2.6851911544799805,
+      "logits/rejected": -2.679932117462158,
+      "logps/chosen": -35.06427001953125,
+      "logps/rejected": -39.153785705566406,
+      "loss": 0.6644,
+      "rewards/accuracies": 0.453125,
+      "rewards/chosen": -0.10253330320119858,
+      "rewards/margins": 0.07344245910644531,
+      "rewards/rejected": -0.1759757697582245,
+      "step": 44
+    },
+    {
+      "epoch": 0.288,
+      "grad_norm": 1.6170339584350586,
+      "learning_rate": 1.6167188726285433e-06,
+      "logits/chosen": -2.6501150131225586,
+      "logits/rejected": -2.6518120765686035,
+      "logps/chosen": -29.792404174804688,
+      "logps/rejected": -32.23724365234375,
+      "loss": 0.6495,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.11595583707094193,
+      "rewards/margins": 0.09647408872842789,
+      "rewards/rejected": -0.2124299257993698,
+      "step": 45
+    },
+    {
+      "epoch": 0.2944,
+      "grad_norm": 1.5615488290786743,
+      "learning_rate": 1.600742264237979e-06,
+      "logits/chosen": -2.588437080383301,
+      "logits/rejected": -2.587428569793701,
+      "logps/chosen": -20.895179748535156,
+      "logps/rejected": -26.258813858032227,
+      "loss": 0.6547,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.08071164041757584,
+      "rewards/margins": 0.08578822016716003,
+      "rewards/rejected": -0.16649985313415527,
+      "step": 46
+    },
+    {
+      "epoch": 0.3008,
+      "grad_norm": 1.6559025049209595,
+      "learning_rate": 1.584522029609889e-06,
+      "logits/chosen": -2.6572139263153076,
+      "logits/rejected": -2.6609270572662354,
+      "logps/chosen": -26.982969284057617,
+      "logps/rejected": -25.127426147460938,
+      "loss": 0.6512,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.08569702506065369,
+      "rewards/margins": 0.09374863654375076,
+      "rewards/rejected": -0.17944566905498505,
+      "step": 47
+    },
+    {
+      "epoch": 0.3072,
+      "grad_norm": 1.5967183113098145,
+      "learning_rate": 1.5680647467311555e-06,
+      "logits/chosen": -2.617936611175537,
+      "logits/rejected": -2.621030330657959,
+      "logps/chosen": -36.36157989501953,
+      "logps/rejected": -38.92307662963867,
+      "loss": 0.6684,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.1475619375705719,
+      "rewards/margins": 0.05436363071203232,
+      "rewards/rejected": -0.20192557573318481,
+      "step": 48
+    },
+    {
+      "epoch": 0.3136,
+      "grad_norm": 1.712814450263977,
+      "learning_rate": 1.5513770897216919e-06,
+      "logits/chosen": -2.603701114654541,
+      "logits/rejected": -2.613720178604126,
+      "logps/chosen": -21.212322235107422,
+      "logps/rejected": -28.460283279418945,
+      "loss": 0.6481,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.08767766505479813,
+      "rewards/margins": 0.0972074642777443,
+      "rewards/rejected": -0.18488512933254242,
+      "step": 49
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 1.4909231662750244,
+      "learning_rate": 1.534465826127801e-06,
+      "logits/chosen": -2.6632070541381836,
+      "logits/rejected": -2.6630711555480957,
+      "logps/chosen": -20.551307678222656,
+      "logps/rejected": -23.05849838256836,
+      "loss": 0.6468,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.07508397847414017,
+      "rewards/margins": 0.10307573527097702,
+      "rewards/rejected": -0.17815972864627838,
+      "step": 50
+    },
+    {
+      "epoch": 0.3264,
+      "grad_norm": 1.6311036348342896,
+      "learning_rate": 1.5173378141776567e-06,
+      "logits/chosen": -2.623863458633423,
+      "logits/rejected": -2.622256278991699,
+      "logps/chosen": -34.566123962402344,
+      "logps/rejected": -36.67254638671875,
+      "loss": 0.6609,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.16467957198619843,
+      "rewards/margins": 0.07420256733894348,
+      "rewards/rejected": -0.2388821542263031,
+      "step": 51
+    },
+    {
+      "epoch": 0.3328,
+      "grad_norm": 1.514613151550293,
+      "learning_rate": 1.5e-06,
+      "logits/chosen": -2.6108975410461426,
+      "logits/rejected": -2.611926794052124,
+      "logps/chosen": -21.698822021484375,
+      "logps/rejected": -23.747154235839844,
+      "loss": 0.6587,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.08456654846668243,
+      "rewards/margins": 0.08089850842952728,
+      "rewards/rejected": -0.16546505689620972,
+      "step": 52
+    },
+    {
+      "epoch": 0.3392,
+      "grad_norm": 1.440454363822937,
+      "learning_rate": 1.4824594148071933e-06,
+      "logits/chosen": -2.6745011806488037,
+      "logits/rejected": -2.674811363220215,
+      "logps/chosen": -23.176433563232422,
+      "logps/rejected": -24.830535888671875,
+      "loss": 0.6477,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.09089958667755127,
+      "rewards/margins": 0.1057419404387474,
+      "rewards/rejected": -0.19664153456687927,
+      "step": 53
+    },
+    {
+      "epoch": 0.3456,
+      "grad_norm": 1.819265604019165,
+      "learning_rate": 1.4647231720437684e-06,
+      "logits/chosen": -2.647128105163574,
+      "logits/rejected": -2.656473159790039,
+      "logps/chosen": -29.09586524963379,
+      "logps/rejected": -34.111854553222656,
+      "loss": 0.6565,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.12301033735275269,
+      "rewards/margins": 0.09625902771949768,
+      "rewards/rejected": -0.21926936507225037,
+      "step": 54
+    },
+    {
+      "epoch": 0.352,
+      "grad_norm": 1.6087982654571533,
+      "learning_rate": 1.4467984645016258e-06,
+      "logits/chosen": -2.6477177143096924,
+      "logits/rejected": -2.6565260887145996,
+      "logps/chosen": -34.84809112548828,
+      "logps/rejected": -37.290138244628906,
+      "loss": 0.6516,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.1687609702348709,
+      "rewards/margins": 0.09631241858005524,
+      "rewards/rejected": -0.26507338881492615,
+      "step": 55
+    },
+    {
+      "epoch": 0.3584,
+      "grad_norm": 1.8970972299575806,
+      "learning_rate": 1.428692561403054e-06,
+      "logits/chosen": -2.665679931640625,
+      "logits/rejected": -2.6677370071411133,
+      "logps/chosen": -24.156490325927734,
+      "logps/rejected": -24.434778213500977,
+      "loss": 0.6524,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.12327785044908524,
+      "rewards/margins": 0.09678448736667633,
+      "rewards/rejected": -0.22006234526634216,
+      "step": 56
+    },
+    {
+      "epoch": 0.3648,
+      "grad_norm": 1.865411639213562,
+      "learning_rate": 1.410412805452757e-06,
+      "logits/chosen": -2.619537591934204,
+      "logits/rejected": -2.620112895965576,
+      "logps/chosen": -24.477069854736328,
+      "logps/rejected": -39.926063537597656,
+      "loss": 0.6166,
+      "rewards/accuracies": 0.671875,
+      "rewards/chosen": -0.09505628049373627,
+      "rewards/margins": 0.18781036138534546,
+      "rewards/rejected": -0.2828666567802429,
+      "step": 57
+    },
+    {
+      "epoch": 0.3712,
+      "grad_norm": 1.5756853818893433,
+      "learning_rate": 1.391966609860075e-06,
+      "logits/chosen": -2.6387054920196533,
+      "logits/rejected": -2.63525390625,
+      "logps/chosen": -28.313343048095703,
+      "logps/rejected": -27.466154098510742,
+      "loss": 0.6464,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.1476786881685257,
+      "rewards/margins": 0.11038099229335785,
+      "rewards/rejected": -0.25805968046188354,
+      "step": 58
+    },
+    {
+      "epoch": 0.3776,
+      "grad_norm": 1.8099141120910645,
+      "learning_rate": 1.3733614553326208e-06,
+      "logits/chosen": -2.6417694091796875,
+      "logits/rejected": -2.644674301147461,
+      "logps/chosen": -35.27565002441406,
+      "logps/rejected": -32.589805603027344,
+      "loss": 0.6744,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.17645558714866638,
+      "rewards/margins": 0.06160591542720795,
+      "rewards/rejected": -0.23806150257587433,
+      "step": 59
+    },
+    {
+      "epoch": 0.384,
+      "grad_norm": 1.7492915391921997,
+      "learning_rate": 1.3546048870425354e-06,
+      "logits/chosen": -2.645029306411743,
+      "logits/rejected": -2.6534972190856934,
+      "logps/chosen": -28.958282470703125,
+      "logps/rejected": -31.320003509521484,
+      "loss": 0.628,
+      "rewards/accuracies": 0.6875,
+      "rewards/chosen": -0.14790010452270508,
+      "rewards/margins": 0.15829242765903473,
+      "rewards/rejected": -0.3061925172805786,
+      "step": 60
+    },
+    {
+      "epoch": 0.3904,
+      "grad_norm": 1.6458604335784912,
+      "learning_rate": 1.3357045115666048e-06,
+      "logits/chosen": -2.642721652984619,
+      "logits/rejected": -2.6435093879699707,
+      "logps/chosen": -23.098867416381836,
+      "logps/rejected": -27.401762008666992,
+      "loss": 0.6312,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.1156662181019783,
+      "rewards/margins": 0.14065474271774292,
+      "rewards/rejected": -0.2563209533691406,
+      "step": 61
+    },
+    {
+      "epoch": 0.3968,
+      "grad_norm": 1.5396157503128052,
+      "learning_rate": 1.3166679938014726e-06,
+      "logits/chosen": -2.6219708919525146,
+      "logits/rejected": -2.6229166984558105,
+      "logps/chosen": -20.742313385009766,
+      "logps/rejected": -24.208255767822266,
+      "loss": 0.6324,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.0953851193189621,
+      "rewards/margins": 0.1443781554698944,
+      "rewards/rejected": -0.2397632598876953,
+      "step": 62
+    },
+    {
+      "epoch": 0.4032,
+      "grad_norm": 1.6560550928115845,
+      "learning_rate": 1.297503053855203e-06,
+      "logits/chosen": -2.602386236190796,
+      "logits/rejected": -2.5967581272125244,
+      "logps/chosen": -24.75385856628418,
+      "logps/rejected": -23.24307632446289,
+      "loss": 0.6679,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.1230187863111496,
+      "rewards/margins": 0.06344311684370041,
+      "rewards/rejected": -0.1864619255065918,
+      "step": 63
+    },
+    {
+      "epoch": 0.4096,
+      "grad_norm": 1.4719752073287964,
+      "learning_rate": 1.2782174639164526e-06,
+      "logits/chosen": -2.6407883167266846,
+      "logits/rejected": -2.641312837600708,
+      "logps/chosen": -22.249114990234375,
+      "logps/rejected": -31.919845581054688,
+      "loss": 0.6445,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.14697568118572235,
+      "rewards/margins": 0.12822286784648895,
+      "rewards/rejected": -0.2751985192298889,
+      "step": 64
+    },
+    {
+      "epoch": 0.416,
+      "grad_norm": 1.6245232820510864,
+      "learning_rate": 1.2588190451025207e-06,
+      "logits/chosen": -2.634258270263672,
+      "logits/rejected": -2.6392617225646973,
+      "logps/chosen": -31.720212936401367,
+      "logps/rejected": -34.2351188659668,
+      "loss": 0.6518,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.16467978060245514,
+      "rewards/margins": 0.09601530432701111,
+      "rewards/rejected": -0.26069507002830505,
+      "step": 65
+    },
+    {
+      "epoch": 0.4224,
+      "grad_norm": 1.751383662223816,
+      "learning_rate": 1.2393156642875577e-06,
+      "logits/chosen": -2.635373592376709,
+      "logits/rejected": -2.6407761573791504,
+      "logps/chosen": -25.238323211669922,
+      "logps/rejected": -30.272274017333984,
+      "loss": 0.621,
+      "rewards/accuracies": 0.6875,
+      "rewards/chosen": -0.11779710650444031,
+      "rewards/margins": 0.17085528373718262,
+      "rewards/rejected": -0.28865236043930054,
+      "step": 66
+    },
+    {
+      "epoch": 0.4288,
+      "grad_norm": 1.7248034477233887,
+      "learning_rate": 1.2197152309122171e-06,
+      "logits/chosen": -2.603132486343384,
+      "logits/rejected": -2.6109507083892822,
+      "logps/chosen": -16.781944274902344,
+      "logps/rejected": -26.596284866333008,
+      "loss": 0.6265,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.09722966700792313,
+      "rewards/margins": 0.16103222966194153,
+      "rewards/rejected": -0.25826191902160645,
+      "step": 67
+    },
+    {
+      "epoch": 0.4352,
+      "grad_norm": 1.5661996603012085,
+      "learning_rate": 1.2000256937760445e-06,
+      "logits/chosen": -2.6702463626861572,
+      "logits/rejected": -2.681889772415161,
+      "logps/chosen": -21.324974060058594,
+      "logps/rejected": -27.644878387451172,
+      "loss": 0.6388,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.15785600244998932,
+      "rewards/margins": 0.1313333511352539,
+      "rewards/rejected": -0.28918933868408203,
+      "step": 68
+    },
+    {
+      "epoch": 0.4416,
+      "grad_norm": 1.5733962059020996,
+      "learning_rate": 1.1802550378139057e-06,
+      "logits/chosen": -2.5827667713165283,
+      "logits/rejected": -2.588782548904419,
+      "logps/chosen": -22.29694175720215,
+      "logps/rejected": -29.593278884887695,
+      "loss": 0.6141,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.1019718125462532,
+      "rewards/margins": 0.18210682272911072,
+      "rewards/rejected": -0.2840786278247833,
+      "step": 69
+    },
+    {
+      "epoch": 0.448,
+      "grad_norm": 1.4367417097091675,
+      "learning_rate": 1.1604112808577602e-06,
+      "logits/chosen": -2.703465461730957,
+      "logits/rejected": -2.702883720397949,
+      "logps/chosen": -18.497364044189453,
+      "logps/rejected": -22.5637264251709,
+      "loss": 0.6388,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.10068204998970032,
+      "rewards/margins": 0.12806305289268494,
+      "rewards/rejected": -0.22874510288238525,
+      "step": 70
+    },
+    {
+      "epoch": 0.4544,
+      "grad_norm": 1.7489335536956787,
+      "learning_rate": 1.1405024703850927e-06,
+      "logits/chosen": -2.6488733291625977,
+      "logits/rejected": -2.65185546875,
+      "logps/chosen": -23.32280921936035,
+      "logps/rejected": -30.010190963745117,
+      "loss": 0.6052,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.11882607638835907,
+      "rewards/margins": 0.20883682370185852,
+      "rewards/rejected": -0.3276628851890564,
+      "step": 71
+    },
+    {
+      "epoch": 0.4608,
+      "grad_norm": 1.4934309720993042,
+      "learning_rate": 1.1205366802553228e-06,
+      "logits/chosen": -2.628666400909424,
+      "logits/rejected": -2.6262712478637695,
+      "logps/chosen": -20.439163208007812,
+      "logps/rejected": -22.379465103149414,
+      "loss": 0.6276,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.13691446185112,
+      "rewards/margins": 0.1575683057308197,
+      "rewards/rejected": -0.2944827675819397,
+      "step": 72
+    },
+    {
+      "epoch": 0.4672,
+      "grad_norm": 1.4480177164077759,
+      "learning_rate": 1.100522007435517e-06,
+      "logits/chosen": -2.6378395557403564,
+      "logits/rejected": -2.6392858028411865,
+      "logps/chosen": -17.709251403808594,
+      "logps/rejected": -21.726078033447266,
+      "loss": 0.6337,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.08252736926078796,
+      "rewards/margins": 0.13930927217006683,
+      "rewards/rejected": -0.2218366414308548,
+      "step": 73
+    },
+    {
+      "epoch": 0.4736,
+      "grad_norm": 1.621299147605896,
+      "learning_rate": 1.080466568716726e-06,
+      "logits/chosen": -2.678636312484741,
+      "logits/rejected": -2.6822118759155273,
+      "logps/chosen": -19.887462615966797,
+      "logps/rejected": -29.38321304321289,
+      "loss": 0.6161,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.14735013246536255,
+      "rewards/margins": 0.19279694557189941,
+      "rewards/rejected": -0.34014707803726196,
+      "step": 74
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 1.523481845855713,
+      "learning_rate": 1.060378497422286e-06,
+      "logits/chosen": -2.6149892807006836,
+      "logits/rejected": -2.6211845874786377,
+      "logps/chosen": -21.412416458129883,
+      "logps/rejected": -23.929367065429688,
+      "loss": 0.6425,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.138633131980896,
+      "rewards/margins": 0.12505067884922028,
+      "rewards/rejected": -0.2636837959289551,
+      "step": 75
+    },
+    {
+      "epoch": 0.4864,
+      "grad_norm": 1.9181902408599854,
+      "learning_rate": 1.0402659401094152e-06,
+      "logits/chosen": -2.648951530456543,
+      "logits/rejected": -2.6508355140686035,
+      "logps/chosen": -32.625492095947266,
+      "logps/rejected": -27.530559539794922,
+      "loss": 0.6687,
+      "rewards/accuracies": 0.5,
+      "rewards/chosen": -0.1979401856660843,
+      "rewards/margins": 0.087333083152771,
+      "rewards/rejected": -0.2852732837200165,
+      "step": 76
+    },
+    {
+      "epoch": 0.4928,
+      "grad_norm": 1.8461145162582397,
+      "learning_rate": 1.0201370532654403e-06,
+      "logits/chosen": -2.5426108837127686,
+      "logits/rejected": -2.546271562576294,
+      "logps/chosen": -20.256982803344727,
+      "logps/rejected": -32.517581939697266,
+      "loss": 0.5872,
+      "rewards/accuracies": 0.734375,
+      "rewards/chosen": -0.12056663632392883,
+      "rewards/margins": 0.2653110921382904,
+      "rewards/rejected": -0.38587772846221924,
+      "step": 77
+    },
+    {
+      "epoch": 0.4992,
+      "grad_norm": 1.6871867179870605,
+      "learning_rate": 1e-06,
+      "logits/chosen": -2.5540127754211426,
+      "logits/rejected": -2.5610508918762207,
+      "logps/chosen": -28.586442947387695,
+      "logps/rejected": -29.418691635131836,
+      "loss": 0.6213,
+      "rewards/accuracies": 0.671875,
+      "rewards/chosen": -0.19238515198230743,
+      "rewards/margins": 0.17244525253772736,
+      "rewards/rejected": -0.3648304045200348,
+      "step": 78
+    },
+    {
+      "epoch": 0.5056,
+      "grad_norm": 1.838136911392212,
+      "learning_rate": 9.798629467345599e-07,
+      "logits/chosen": -2.6480908393859863,
+      "logits/rejected": -2.6476452350616455,
+      "logps/chosen": -33.84926223754883,
+      "logps/rejected": -37.04065704345703,
+      "loss": 0.655,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.2544393241405487,
+      "rewards/margins": 0.1295141577720642,
+      "rewards/rejected": -0.3839534521102905,
+      "step": 79
+    },
+    {
+      "epoch": 0.512,
+      "grad_norm": 1.767898678779602,
+      "learning_rate": 9.597340598905851e-07,
+      "logits/chosen": -2.6242833137512207,
+      "logits/rejected": -2.6161022186279297,
+      "logps/chosen": -33.3975830078125,
+      "logps/rejected": -38.89872741699219,
+      "loss": 0.6179,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.21687382459640503,
+      "rewards/margins": 0.2002078890800476,
+      "rewards/rejected": -0.417081743478775,
+      "step": 80
+    },
+    {
+      "epoch": 0.5184,
+      "grad_norm": 1.5765364170074463,
+      "learning_rate": 9.396215025777138e-07,
+      "logits/chosen": -2.627566337585449,
+      "logits/rejected": -2.6267096996307373,
+      "logps/chosen": -27.6579532623291,
+      "logps/rejected": -29.39544677734375,
+      "loss": 0.6386,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.17304903268814087,
+      "rewards/margins": 0.1442856341600418,
+      "rewards/rejected": -0.3173346519470215,
+      "step": 81
+    },
+    {
+      "epoch": 0.5248,
+      "grad_norm": 1.5013419389724731,
+      "learning_rate": 9.195334312832741e-07,
+      "logits/chosen": -2.6478822231292725,
+      "logits/rejected": -2.652007818222046,
+      "logps/chosen": -20.743148803710938,
+      "logps/rejected": -27.11212158203125,
+      "loss": 0.6184,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.12529368698596954,
+      "rewards/margins": 0.1996924877166748,
+      "rewards/rejected": -0.32498618960380554,
+      "step": 82
+    },
+    {
+      "epoch": 0.5312,
+      "grad_norm": 2.5118231773376465,
+      "learning_rate": 8.99477992564483e-07,
+      "logits/chosen": -2.547119617462158,
+      "logits/rejected": -2.547783136367798,
+      "logps/chosen": -33.352935791015625,
+      "logps/rejected": -26.84307861328125,
+      "loss": 0.653,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.22605080902576447,
+      "rewards/margins": 0.14342014491558075,
+      "rewards/rejected": -0.3694709539413452,
+      "step": 83
+    },
+    {
+      "epoch": 0.5376,
+      "grad_norm": 1.9225335121154785,
+      "learning_rate": 8.79463319744677e-07,
+      "logits/chosen": -2.59800124168396,
+      "logits/rejected": -2.608645439147949,
+      "logps/chosen": -21.690284729003906,
+      "logps/rejected": -31.604402542114258,
+      "loss": 0.6248,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.16822285950183868,
+      "rewards/margins": 0.21106992661952972,
+      "rewards/rejected": -0.3792927861213684,
+      "step": 84
+    },
+    {
+      "epoch": 0.544,
+      "grad_norm": 1.4944307804107666,
+      "learning_rate": 8.594975296149075e-07,
+      "logits/chosen": -2.625525951385498,
+      "logits/rejected": -2.6290879249572754,
+      "logps/chosen": -20.582530975341797,
+      "logps/rejected": -26.895797729492188,
+      "loss": 0.5949,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.12493821978569031,
+      "rewards/margins": 0.25203022360801697,
+      "rewards/rejected": -0.3769684433937073,
+      "step": 85
+    },
+    {
+      "epoch": 0.5504,
+      "grad_norm": 1.5492219924926758,
+      "learning_rate": 8.395887191422396e-07,
+      "logits/chosen": -2.6404764652252197,
+      "logits/rejected": -2.641653537750244,
+      "logps/chosen": -22.250150680541992,
+      "logps/rejected": -24.068775177001953,
+      "loss": 0.6498,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.1951374113559723,
+      "rewards/margins": 0.11561567336320877,
+      "rewards/rejected": -0.31075307726860046,
+      "step": 86
+    },
+    {
+      "epoch": 0.5568,
+      "grad_norm": 1.5819690227508545,
+      "learning_rate": 8.197449621860942e-07,
+      "logits/chosen": -2.5950114727020264,
+      "logits/rejected": -2.596634864807129,
+      "logps/chosen": -21.68282699584961,
+      "logps/rejected": -26.268205642700195,
+      "loss": 0.6201,
+      "rewards/accuracies": 0.671875,
+      "rewards/chosen": -0.18349911272525787,
+      "rewards/margins": 0.1877526342868805,
+      "rewards/rejected": -0.3712517321109772,
+      "step": 87
+    },
+    {
+      "epoch": 0.5632,
+      "grad_norm": 1.7474534511566162,
+      "learning_rate": 7.999743062239557e-07,
+      "logits/chosen": -2.6187121868133545,
+      "logits/rejected": -2.611988067626953,
+      "logps/chosen": -24.89275360107422,
+      "logps/rejected": -30.300533294677734,
+      "loss": 0.6129,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.17396652698516846,
+      "rewards/margins": 0.22732898592948914,
+      "rewards/rejected": -0.40129554271698,
+      "step": 88
+    },
+    {
+      "epoch": 0.5696,
+      "grad_norm": 1.6722171306610107,
+      "learning_rate": 7.802847690877831e-07,
+      "logits/chosen": -2.604179859161377,
+      "logits/rejected": -2.6083335876464844,
+      "logps/chosen": -23.76987075805664,
+      "logps/rejected": -34.22930908203125,
+      "loss": 0.6035,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.20327410101890564,
+      "rewards/margins": 0.2363450527191162,
+      "rewards/rejected": -0.43961915373802185,
+      "step": 89
+    },
+    {
+      "epoch": 0.576,
+      "grad_norm": 1.8027116060256958,
+      "learning_rate": 7.606843357124425e-07,
+      "logits/chosen": -2.60955810546875,
+      "logits/rejected": -2.6146914958953857,
+      "logps/chosen": -20.799884796142578,
+      "logps/rejected": -26.12914276123047,
+      "loss": 0.6036,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.13158757984638214,
+      "rewards/margins": 0.2140113115310669,
+      "rewards/rejected": -0.34559887647628784,
+      "step": 90
+    },
+    {
+      "epoch": 0.5824,
+      "grad_norm": 1.5857772827148438,
+      "learning_rate": 7.411809548974791e-07,
+      "logits/chosen": -2.59030818939209,
+      "logits/rejected": -2.5919747352600098,
+      "logps/chosen": -23.269086837768555,
+      "logps/rejected": -26.460769653320312,
+      "loss": 0.6367,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.15069209039211273,
+      "rewards/margins": 0.14479246735572815,
+      "rewards/rejected": -0.2954845428466797,
+      "step": 91
+    },
+    {
+      "epoch": 0.5888,
+      "grad_norm": 1.6919114589691162,
+      "learning_rate": 7.217825360835473e-07,
+      "logits/chosen": -2.6045777797698975,
+      "logits/rejected": -2.6080751419067383,
+      "logps/chosen": -28.70133399963379,
+      "logps/rejected": -37.94871139526367,
+      "loss": 0.5995,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.17101868987083435,
+      "rewards/margins": 0.25572121143341064,
+      "rewards/rejected": -0.4267398715019226,
+      "step": 92
+    },
+    {
+      "epoch": 0.5952,
+      "grad_norm": 1.5466245412826538,
+      "learning_rate": 7.024969461447972e-07,
+      "logits/chosen": -2.6184473037719727,
+      "logits/rejected": -2.61545729637146,
+      "logps/chosen": -27.17377471923828,
+      "logps/rejected": -31.0013484954834,
+      "loss": 0.6456,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.19300419092178345,
+      "rewards/margins": 0.1320933848619461,
+      "rewards/rejected": -0.32509759068489075,
+      "step": 93
+    },
+    {
+      "epoch": 0.6016,
+      "grad_norm": 1.749213457107544,
+      "learning_rate": 6.833320061985277e-07,
+      "logits/chosen": -2.6235976219177246,
+      "logits/rejected": -2.6337835788726807,
+      "logps/chosen": -30.744365692138672,
+      "logps/rejected": -31.844709396362305,
+      "loss": 0.6351,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.20801806449890137,
+      "rewards/margins": 0.1620381474494934,
+      "rewards/rejected": -0.3700562119483948,
+      "step": 94
+    },
+    {
+      "epoch": 0.608,
+      "grad_norm": 1.4693808555603027,
+      "learning_rate": 6.642954884333954e-07,
+      "logits/chosen": -2.6233415603637695,
+      "logits/rejected": -2.619433879852295,
+      "logps/chosen": -26.5659122467041,
+      "logps/rejected": -38.52685546875,
+      "loss": 0.6033,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.1947498917579651,
+      "rewards/margins": 0.25007978081703186,
+      "rewards/rejected": -0.44482964277267456,
+      "step": 95
+    },
+    {
+      "epoch": 0.6144,
+      "grad_norm": 1.5889641046524048,
+      "learning_rate": 6.453951129574643e-07,
+      "logits/chosen": -2.6244001388549805,
+      "logits/rejected": -2.630420207977295,
+      "logps/chosen": -24.560443878173828,
+      "logps/rejected": -26.42283058166504,
+      "loss": 0.6547,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.18999485671520233,
+      "rewards/margins": 0.10254239290952682,
+      "rewards/rejected": -0.29253727197647095,
+      "step": 96
+    },
+    {
+      "epoch": 0.6208,
+      "grad_norm": 1.6251789331436157,
+      "learning_rate": 6.26638544667379e-07,
+      "logits/chosen": -2.5631370544433594,
+      "logits/rejected": -2.5691609382629395,
+      "logps/chosen": -33.219966888427734,
+      "logps/rejected": -40.98374938964844,
+      "loss": 0.5912,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.2685796618461609,
+      "rewards/margins": 0.26844310760498047,
+      "rewards/rejected": -0.5370227694511414,
+      "step": 97
+    },
+    {
+      "epoch": 0.6272,
+      "grad_norm": 1.6536494493484497,
+      "learning_rate": 6.08033390139925e-07,
+      "logits/chosen": -2.612924575805664,
+      "logits/rejected": -2.6108927726745605,
+      "logps/chosen": -30.917957305908203,
+      "logps/rejected": -33.09983444213867,
+      "loss": 0.6404,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.24251434206962585,
+      "rewards/margins": 0.14115232229232788,
+      "rewards/rejected": -0.3836666941642761,
+      "step": 98
+    },
+    {
+      "epoch": 0.6336,
+      "grad_norm": 1.4068188667297363,
+      "learning_rate": 5.895871945472434e-07,
+      "logits/chosen": -2.6371684074401855,
+      "logits/rejected": -2.6351089477539062,
+      "logps/chosen": -27.32377815246582,
+      "logps/rejected": -29.539642333984375,
+      "loss": 0.6279,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.21430930495262146,
+      "rewards/margins": 0.1696455478668213,
+      "rewards/rejected": -0.38395485281944275,
+      "step": 99
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 1.474990725517273,
+      "learning_rate": 5.713074385969456e-07,
+      "logits/chosen": -2.585543632507324,
+      "logits/rejected": -2.588040828704834,
+      "logps/chosen": -19.979612350463867,
+      "logps/rejected": -23.094497680664062,
+      "loss": 0.6504,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.1839856207370758,
+      "rewards/margins": 0.11544550955295563,
+      "rewards/rejected": -0.2994311451911926,
+      "step": 100
+    },
+    {
+      "epoch": 0.6464,
+      "grad_norm": 2.0331740379333496,
+      "learning_rate": 5.532015354983741e-07,
+      "logits/chosen": -2.632883071899414,
+      "logits/rejected": -2.6397085189819336,
+      "logps/chosen": -25.389251708984375,
+      "logps/rejected": -28.902347564697266,
+      "loss": 0.6508,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.2711215019226074,
+      "rewards/margins": 0.15430796146392822,
+      "rewards/rejected": -0.42542946338653564,
+      "step": 101
+    },
+    {
+      "epoch": 0.6528,
+      "grad_norm": 1.8769034147262573,
+      "learning_rate": 5.352768279562314e-07,
+      "logits/chosen": -2.600227117538452,
+      "logits/rejected": -2.589409589767456,
+      "logps/chosen": -26.329830169677734,
+      "logps/rejected": -30.269969940185547,
+      "loss": 0.6362,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.22971783578395844,
+      "rewards/margins": 0.1924903690814972,
+      "rewards/rejected": -0.42220818996429443,
+      "step": 102
+    },
+    {
+      "epoch": 0.6592,
+      "grad_norm": 2.3717446327209473,
+      "learning_rate": 5.175405851928068e-07,
+      "logits/chosen": -2.6390957832336426,
+      "logits/rejected": -2.6411001682281494,
+      "logps/chosen": -28.72641944885254,
+      "logps/rejected": -23.84124755859375,
+      "loss": 0.6737,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.18652591109275818,
+      "rewards/margins": 0.07918506860733032,
+      "rewards/rejected": -0.2657109797000885,
+      "step": 103
+    },
+    {
+      "epoch": 0.6656,
+      "grad_norm": 1.7093194723129272,
+      "learning_rate": 5.000000000000002e-07,
+      "logits/chosen": -2.6221108436584473,
+      "logits/rejected": -2.6264395713806152,
+      "logps/chosen": -23.432836532592773,
+      "logps/rejected": -35.811988830566406,
+      "loss": 0.578,
+      "rewards/accuracies": 0.734375,
+      "rewards/chosen": -0.15505719184875488,
+      "rewards/margins": 0.29802069067955017,
+      "rewards/rejected": -0.45307785272598267,
+      "step": 104
+    },
+    {
+      "epoch": 0.672,
+      "grad_norm": 2.126798152923584,
+      "learning_rate": 4.826621858223431e-07,
+      "logits/chosen": -2.6108996868133545,
+      "logits/rejected": -2.599609375,
+      "logps/chosen": -35.375606536865234,
+      "logps/rejected": -28.418149948120117,
+      "loss": 0.671,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.2683427333831787,
+      "rewards/margins": 0.08595550805330276,
+      "rewards/rejected": -0.3542982339859009,
+      "step": 105
+    },
+    {
+      "epoch": 0.6784,
+      "grad_norm": 1.6522189378738403,
+      "learning_rate": 4.6553417387219884e-07,
+      "logits/chosen": -2.6088504791259766,
+      "logits/rejected": -2.6069254875183105,
+      "logps/chosen": -27.17560577392578,
+      "logps/rejected": -31.48743438720703,
+      "loss": 0.6379,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.208372563123703,
+      "rewards/margins": 0.14145778119564056,
+      "rewards/rejected": -0.34983035922050476,
+      "step": 106
+    },
+    {
+      "epoch": 0.6848,
+      "grad_norm": 1.4990570545196533,
+      "learning_rate": 4.486229102783083e-07,
+      "logits/chosen": -2.6148300170898438,
+      "logits/rejected": -2.6092538833618164,
+      "logps/chosen": -26.90603256225586,
+      "logps/rejected": -34.3206787109375,
+      "loss": 0.6079,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.23308473825454712,
+      "rewards/margins": 0.23289112746715546,
+      "rewards/rejected": -0.46597588062286377,
+      "step": 107
+    },
+    {
+      "epoch": 0.6912,
+      "grad_norm": 1.8453425168991089,
+      "learning_rate": 4.3193525326884426e-07,
+      "logits/chosen": -2.5528130531311035,
+      "logits/rejected": -2.547475814819336,
+      "logps/chosen": -32.27470779418945,
+      "logps/rejected": -45.245567321777344,
+      "loss": 0.6143,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.2673153877258301,
+      "rewards/margins": 0.2748442590236664,
+      "rewards/rejected": -0.5421596765518188,
+      "step": 108
+    },
+    {
+      "epoch": 0.6976,
+      "grad_norm": 1.5140947103500366,
+      "learning_rate": 4.1547797039011135e-07,
+      "logits/chosen": -2.602471113204956,
+      "logits/rejected": -2.598973035812378,
+      "logps/chosen": -29.44887924194336,
+      "logps/rejected": -36.79751968383789,
+      "loss": 0.5908,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.22240225970745087,
+      "rewards/margins": 0.2742350101470947,
+      "rewards/rejected": -0.4966372847557068,
+      "step": 109
+    },
+    {
+      "epoch": 0.704,
+      "grad_norm": 1.535190224647522,
+      "learning_rate": 3.9925773576202094e-07,
+      "logits/chosen": -2.599069118499756,
+      "logits/rejected": -2.598602294921875,
+      "logps/chosen": -26.50482940673828,
+      "logps/rejected": -28.245283126831055,
+      "loss": 0.617,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.216806560754776,
+      "rewards/margins": 0.19334101676940918,
+      "rewards/rejected": -0.4101475775241852,
+      "step": 110
+    },
+    {
+      "epoch": 0.7104,
+      "grad_norm": 1.744052767753601,
+      "learning_rate": 3.832811273714569e-07,
+      "logits/chosen": -2.585948944091797,
+      "logits/rejected": -2.5806031227111816,
+      "logps/chosen": -23.81147003173828,
+      "logps/rejected": -26.963855743408203,
+      "loss": 0.6445,
+      "rewards/accuracies": 0.5,
+      "rewards/chosen": -0.1732327789068222,
+      "rewards/margins": 0.14166364073753357,
+      "rewards/rejected": -0.3148964047431946,
+      "step": 111
+    },
+    {
+      "epoch": 0.7168,
+      "grad_norm": 1.695529818534851,
+      "learning_rate": 3.675546244046228e-07,
+      "logits/chosen": -2.6089272499084473,
+      "logits/rejected": -2.616572380065918,
+      "logps/chosen": -23.279071807861328,
+      "logps/rejected": -25.473308563232422,
+      "loss": 0.6271,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.14990699291229248,
+      "rewards/margins": 0.18665853142738342,
+      "rewards/rejected": -0.3365654945373535,
+      "step": 112
+    },
+    {
+      "epoch": 0.7232,
+      "grad_norm": 1.6549317836761475,
+      "learning_rate": 3.5208460461946133e-07,
+      "logits/chosen": -2.6233081817626953,
+      "logits/rejected": -2.6279563903808594,
+      "logps/chosen": -26.11295509338379,
+      "logps/rejected": -31.000097274780273,
+      "loss": 0.6404,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.2790622115135193,
+      "rewards/margins": 0.15750598907470703,
+      "rewards/rejected": -0.4365682303905487,
+      "step": 113
+    },
+    {
+      "epoch": 0.7296,
+      "grad_norm": 1.6121848821640015,
+      "learning_rate": 3.3687734175920503e-07,
+      "logits/chosen": -2.5145516395568848,
+      "logits/rejected": -2.5188541412353516,
+      "logps/chosen": -22.996007919311523,
+      "logps/rejected": -35.4892692565918,
+      "loss": 0.6044,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.18752524256706238,
+      "rewards/margins": 0.25176137685775757,
+      "rewards/rejected": -0.43928664922714233,
+      "step": 114
+    },
+    {
+      "epoch": 0.736,
+      "grad_norm": 1.8857587575912476,
+      "learning_rate": 3.2193900300810904e-07,
+      "logits/chosen": -2.6584115028381348,
+      "logits/rejected": -2.662964344024658,
+      "logps/chosen": -24.670669555664062,
+      "logps/rejected": -31.12428092956543,
+      "loss": 0.5836,
+      "rewards/accuracies": 0.671875,
+      "rewards/chosen": -0.14637771248817444,
+      "rewards/margins": 0.29260551929473877,
+      "rewards/rejected": -0.4389832019805908,
+      "step": 115
+    },
+    {
+      "epoch": 0.7424,
+      "grad_norm": 1.5393645763397217,
+      "learning_rate": 3.072756464904006e-07,
+      "logits/chosen": -2.5413527488708496,
+      "logits/rejected": -2.541822910308838,
+      "logps/chosen": -25.6234130859375,
+      "logps/rejected": -33.03318786621094,
+      "loss": 0.6137,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.21068117022514343,
+      "rewards/margins": 0.21521715819835663,
+      "rewards/rejected": -0.42589831352233887,
+      "step": 116
+    },
+    {
+      "epoch": 0.7488,
+      "grad_norm": 1.7721267938613892,
+      "learning_rate": 2.9289321881345254e-07,
+      "logits/chosen": -2.6014585494995117,
+      "logits/rejected": -2.602719783782959,
+      "logps/chosen": -24.250324249267578,
+      "logps/rejected": -28.464712142944336,
+      "loss": 0.6225,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.19951701164245605,
+      "rewards/margins": 0.1980157047510147,
+      "rewards/rejected": -0.39753273129463196,
+      "step": 117
+    },
+    {
+      "epoch": 0.7552,
+      "grad_norm": 1.5414998531341553,
+      "learning_rate": 2.7879755265618557e-07,
+      "logits/chosen": -2.5935285091400146,
+      "logits/rejected": -2.596818685531616,
+      "logps/chosen": -22.548786163330078,
+      "logps/rejected": -25.119014739990234,
+      "loss": 0.597,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.14906607568264008,
+      "rewards/margins": 0.2550210952758789,
+      "rewards/rejected": -0.4040871858596802,
+      "step": 118
+    },
+    {
+      "epoch": 0.7616,
+      "grad_norm": 1.5603265762329102,
+      "learning_rate": 2.649943644036716e-07,
+      "logits/chosen": -2.5638105869293213,
+      "logits/rejected": -2.560965061187744,
+      "logps/chosen": -22.765018463134766,
+      "logps/rejected": -24.522743225097656,
+      "loss": 0.6394,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.2109951227903366,
+      "rewards/margins": 0.14377270638942719,
+      "rewards/rejected": -0.3547678291797638,
+      "step": 119
+    },
+    {
+      "epoch": 0.768,
+      "grad_norm": 1.9955023527145386,
+      "learning_rate": 2.514892518288988e-07,
+      "logits/chosen": -2.603156328201294,
+      "logits/rejected": -2.597612142562866,
+      "logps/chosen": -26.43991470336914,
+      "logps/rejected": -25.16839599609375,
+      "loss": 0.6675,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.25359681248664856,
+      "rewards/margins": 0.11165492236614227,
+      "rewards/rejected": -0.365251749753952,
+      "step": 120
+    },
+    {
+      "epoch": 0.7744,
+      "grad_norm": 1.5535807609558105,
+      "learning_rate": 2.3828769182264085e-07,
+      "logits/chosen": -2.634467840194702,
+      "logits/rejected": -2.630577564239502,
+      "logps/chosen": -22.262788772583008,
+      "logps/rejected": -21.859067916870117,
+      "loss": 0.6162,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.13089901208877563,
+      "rewards/margins": 0.1952367126941681,
+      "rewards/rejected": -0.3261357545852661,
+      "step": 121
+    },
+    {
+      "epoch": 0.7808,
+      "grad_norm": 1.646085500717163,
+      "learning_rate": 2.253950381723455e-07,
+      "logits/chosen": -2.594827175140381,
+      "logits/rejected": -2.598013401031494,
+      "logps/chosen": -20.885169982910156,
+      "logps/rejected": -33.953399658203125,
+      "loss": 0.6145,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.19412222504615784,
+      "rewards/margins": 0.2344159632921219,
+      "rewards/rejected": -0.4285382032394409,
+      "step": 122
+    },
+    {
+      "epoch": 0.7872,
+      "grad_norm": 1.7772481441497803,
+      "learning_rate": 2.128165193909499e-07,
+      "logits/chosen": -2.5993332862854004,
+      "logits/rejected": -2.60774827003479,
+      "logps/chosen": -23.634796142578125,
+      "logps/rejected": -24.75752067565918,
+      "loss": 0.619,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.19775912165641785,
+      "rewards/margins": 0.20522838830947876,
+      "rewards/rejected": -0.4029875099658966,
+      "step": 123
+    },
+    {
+      "epoch": 0.7936,
+      "grad_norm": 1.7965983152389526,
+      "learning_rate": 2.0055723659649905e-07,
+      "logits/chosen": -2.6080384254455566,
+      "logits/rejected": -2.6146090030670166,
+      "logps/chosen": -22.636978149414062,
+      "logps/rejected": -24.64559555053711,
+      "loss": 0.6448,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.2306388020515442,
+      "rewards/margins": 0.14704349637031555,
+      "rewards/rejected": -0.37768229842185974,
+      "step": 124
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 2.157348155975342,
+      "learning_rate": 1.8862216144342692e-07,
+      "logits/chosen": -2.573525905609131,
+      "logits/rejected": -2.5768914222717285,
+      "logps/chosen": -34.6945915222168,
+      "logps/rejected": -29.599353790283203,
+      "loss": 0.6464,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.26834872364997864,
+      "rewards/margins": 0.13867738842964172,
+      "rewards/rejected": -0.40702611207962036,
+      "step": 125
+    },
+    {
+      "epoch": 0.8064,
+      "grad_norm": 1.62916898727417,
+      "learning_rate": 1.7701613410634365e-07,
+      "logits/chosen": -2.6236939430236816,
+      "logits/rejected": -2.627351999282837,
+      "logps/chosen": -23.213960647583008,
+      "logps/rejected": -34.45604705810547,
+      "loss": 0.5838,
+      "rewards/accuracies": 0.671875,
+      "rewards/chosen": -0.17064885795116425,
+      "rewards/margins": 0.301230788230896,
+      "rewards/rejected": -0.47187966108322144,
+      "step": 126
+    },
+    {
+      "epoch": 0.8128,
+      "grad_norm": 2.014199733734131,
+      "learning_rate": 1.657438613171387e-07,
+      "logits/chosen": -2.6167235374450684,
+      "logits/rejected": -2.6315553188323975,
+      "logps/chosen": -26.410276412963867,
+      "logps/rejected": -31.544479370117188,
+      "loss": 0.5897,
+      "rewards/accuracies": 0.703125,
+      "rewards/chosen": -0.18312443792819977,
+      "rewards/margins": 0.2821982204914093,
+      "rewards/rejected": -0.4653226435184479,
+      "step": 127
+    },
+    {
+      "epoch": 0.8192,
+      "grad_norm": 1.9933661222457886,
+      "learning_rate": 1.5480991445620538e-07,
+      "logits/chosen": -2.6267199516296387,
+      "logits/rejected": -2.6314663887023926,
+      "logps/chosen": -31.201932907104492,
+      "logps/rejected": -32.41779327392578,
+      "loss": 0.6376,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.22118794918060303,
+      "rewards/margins": 0.18489961326122284,
+      "rewards/rejected": -0.4060875475406647,
+      "step": 128
+    },
+    {
+      "epoch": 0.8256,
+      "grad_norm": 2.2170119285583496,
+      "learning_rate": 1.442187276985526e-07,
+      "logits/chosen": -2.5758519172668457,
+      "logits/rejected": -2.582484245300293,
+      "logps/chosen": -26.29662322998047,
+      "logps/rejected": -27.841602325439453,
+      "loss": 0.6449,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.2184506058692932,
+      "rewards/margins": 0.175899475812912,
+      "rewards/rejected": -0.3943500816822052,
+      "step": 129
+    },
+    {
+      "epoch": 0.832,
+      "grad_norm": 1.5597612857818604,
+      "learning_rate": 1.3397459621556128e-07,
+      "logits/chosen": -2.57673978805542,
+      "logits/rejected": -2.5750584602355957,
+      "logps/chosen": -29.762256622314453,
+      "logps/rejected": -41.444793701171875,
+      "loss": 0.6029,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.24154219031333923,
+      "rewards/margins": 0.23838648200035095,
+      "rewards/rejected": -0.4799287021160126,
+      "step": 130
+    },
+    {
+      "epoch": 0.8384,
+      "grad_norm": 2.817040205001831,
+      "learning_rate": 1.2408167443311213e-07,
+      "logits/chosen": -2.5604448318481445,
+      "logits/rejected": -2.5600619316101074,
+      "logps/chosen": -29.55309295654297,
+      "logps/rejected": -25.836933135986328,
+      "loss": 0.6666,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.24983759224414825,
+      "rewards/margins": 0.16183120012283325,
+      "rewards/rejected": -0.4116687774658203,
+      "step": 131
+    },
+    {
+      "epoch": 0.8448,
+      "grad_norm": 1.6076332330703735,
+      "learning_rate": 1.145439743467902e-07,
+      "logits/chosen": -2.5236124992370605,
+      "logits/rejected": -2.5241150856018066,
+      "logps/chosen": -24.745330810546875,
+      "logps/rejected": -27.36791229248047,
+      "loss": 0.6471,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.21809431910514832,
+      "rewards/margins": 0.12445776164531708,
+      "rewards/rejected": -0.3425520658493042,
+      "step": 132
+    },
+    {
+      "epoch": 0.8512,
+      "grad_norm": 1.8516141176223755,
+      "learning_rate": 1.0536536389485273e-07,
+      "logits/chosen": -2.60581636428833,
+      "logits/rejected": -2.6185266971588135,
+      "logps/chosen": -24.80894660949707,
+      "logps/rejected": -28.902212142944336,
+      "loss": 0.6142,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.2200690507888794,
+      "rewards/margins": 0.20549264550209045,
+      "rewards/rejected": -0.42556172609329224,
+      "step": 133
+    },
+    {
+      "epoch": 0.8576,
+      "grad_norm": 2.0396194458007812,
+      "learning_rate": 9.65495653896179e-08,
+      "logits/chosen": -2.6002583503723145,
+      "logits/rejected": -2.6121134757995605,
+      "logps/chosen": -29.090770721435547,
+      "logps/rejected": -34.06471252441406,
+      "loss": 0.589,
+      "rewards/accuracies": 0.671875,
+      "rewards/chosen": -0.1933280974626541,
+      "rewards/margins": 0.29279816150665283,
+      "rewards/rejected": -0.4861263036727905,
+      "step": 134
+    },
+    {
+      "epoch": 0.864,
+      "grad_norm": 1.813738226890564,
+      "learning_rate": 8.810015400790993e-08,
+      "logits/chosen": -2.5964789390563965,
+      "logits/rejected": -2.5941343307495117,
+      "logps/chosen": -22.755495071411133,
+      "logps/rejected": -28.09613609313965,
+      "loss": 0.5606,
+      "rewards/accuracies": 0.703125,
+      "rewards/chosen": -0.14621831476688385,
+      "rewards/margins": 0.3352219760417938,
+      "rewards/rejected": -0.4814402759075165,
+      "step": 135
+    },
+    {
+      "epoch": 0.8704,
+      "grad_norm": 1.6616170406341553,
+      "learning_rate": 8.002055634117577e-08,
+      "logits/chosen": -2.626329183578491,
+      "logits/rejected": -2.6256604194641113,
+      "logps/chosen": -22.777183532714844,
+      "logps/rejected": -26.27685546875,
+      "loss": 0.5921,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.18743543326854706,
+      "rewards/margins": 0.2601334750652313,
+      "rewards/rejected": -0.4475689232349396,
+      "step": 136
+    },
+    {
+      "epoch": 0.8768,
+      "grad_norm": 1.9431946277618408,
+      "learning_rate": 7.231404900585714e-08,
+      "logits/chosen": -2.6369595527648926,
+      "logits/rejected": -2.6498022079467773,
+      "logps/chosen": -30.607051849365234,
+      "logps/rejected": -33.8561897277832,
+      "loss": 0.5875,
+      "rewards/accuracies": 0.734375,
+      "rewards/chosen": -0.21680429577827454,
+      "rewards/margins": 0.2610439658164978,
+      "rewards/rejected": -0.47784826159477234,
+      "step": 137
+    },
+    {
+      "epoch": 0.8832,
+      "grad_norm": 1.634878158569336,
+      "learning_rate": 6.498375731458527e-08,
+      "logits/chosen": -2.603696584701538,
+      "logits/rejected": -2.6012911796569824,
+      "logps/chosen": -23.165857315063477,
+      "logps/rejected": -32.97418975830078,
+      "loss": 0.5893,
+      "rewards/accuracies": 0.671875,
+      "rewards/chosen": -0.18626856803894043,
+      "rewards/margins": 0.2745392322540283,
+      "rewards/rejected": -0.46080780029296875,
+      "step": 138
+    },
+    {
+      "epoch": 0.8896,
+      "grad_norm": 2.0545766353607178,
+      "learning_rate": 5.8032654008735136e-08,
+      "logits/chosen": -2.626938819885254,
+      "logits/rejected": -2.6309990882873535,
+      "logps/chosen": -26.33625030517578,
+      "logps/rejected": -26.596893310546875,
+      "loss": 0.6445,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.2268519401550293,
+      "rewards/margins": 0.16276869177818298,
+      "rewards/rejected": -0.3896206319332123,
+      "step": 139
+    },
+    {
+      "epoch": 0.896,
+      "grad_norm": 1.6766297817230225,
+      "learning_rate": 5.146355805285451e-08,
+      "logits/chosen": -2.6150403022766113,
+      "logits/rejected": -2.614422559738159,
+      "logps/chosen": -22.33306121826172,
+      "logps/rejected": -28.856403350830078,
+      "loss": 0.5961,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.19910869002342224,
+      "rewards/margins": 0.2571624517440796,
+      "rewards/rejected": -0.45627111196517944,
+      "step": 140
+    },
+    {
+      "epoch": 0.9024,
+      "grad_norm": 1.6276016235351562,
+      "learning_rate": 4.52791334914544e-08,
+      "logits/chosen": -2.5284605026245117,
+      "logits/rejected": -2.534757375717163,
+      "logps/chosen": -30.55624008178711,
+      "logps/rejected": -39.18238067626953,
+      "loss": 0.5806,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.25601306557655334,
+      "rewards/margins": 0.301752507686615,
+      "rewards/rejected": -0.5577656030654907,
+      "step": 141
+    },
+    {
+      "epoch": 0.9088,
+      "grad_norm": 1.8015527725219727,
+      "learning_rate": 3.948188836862776e-08,
+      "logits/chosen": -2.611663341522217,
+      "logits/rejected": -2.617194652557373,
+      "logps/chosen": -32.89912033081055,
+      "logps/rejected": -34.12856674194336,
+      "loss": 0.5824,
+      "rewards/accuracies": 0.703125,
+      "rewards/chosen": -0.21851831674575806,
+      "rewards/margins": 0.28189122676849365,
+      "rewards/rejected": -0.5004095435142517,
+      "step": 142
+    },
+    {
+      "epoch": 0.9152,
+      "grad_norm": 1.5642485618591309,
+      "learning_rate": 3.4074173710931796e-08,
+      "logits/chosen": -2.60264253616333,
+      "logits/rejected": -2.603081226348877,
+      "logps/chosen": -21.82186508178711,
+      "logps/rejected": -27.256336212158203,
+      "loss": 0.6123,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.19482672214508057,
+      "rewards/margins": 0.21886488795280457,
+      "rewards/rejected": -0.41369161009788513,
+      "step": 143
+    },
+    {
+      "epoch": 0.9216,
+      "grad_norm": 1.7585622072219849,
+      "learning_rate": 2.9058182573947986e-08,
+      "logits/chosen": -2.6364738941192627,
+      "logits/rejected": -2.6343765258789062,
+      "logps/chosen": -25.372947692871094,
+      "logps/rejected": -27.278942108154297,
+      "loss": 0.6129,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.19601738452911377,
+      "rewards/margins": 0.20809179544448853,
+      "rewards/rejected": -0.4041092097759247,
+      "step": 144
+    },
+    {
+      "epoch": 0.928,
+      "grad_norm": 1.5573078393936157,
+      "learning_rate": 2.4435949152906144e-08,
+      "logits/chosen": -2.547656774520874,
+      "logits/rejected": -2.5421080589294434,
+      "logps/chosen": -30.91408348083496,
+      "logps/rejected": -40.54969787597656,
+      "loss": 0.5799,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": -0.25765368342399597,
+      "rewards/margins": 0.3290165066719055,
+      "rewards/rejected": -0.5866702198982239,
+      "step": 145
+    },
+    {
+      "epoch": 0.9344,
+      "grad_norm": 1.6278663873672485,
+      "learning_rate": 2.0209347957732326e-08,
+      "logits/chosen": -2.639927387237549,
+      "logits/rejected": -2.643014430999756,
+      "logps/chosen": -19.10517692565918,
+      "logps/rejected": -29.58492088317871,
+      "loss": 0.6297,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.18636533617973328,
+      "rewards/margins": 0.17429682612419128,
+      "rewards/rejected": -0.36066216230392456,
+      "step": 146
+    },
+    {
+      "epoch": 0.9408,
+      "grad_norm": 1.7673671245574951,
+      "learning_rate": 1.6380093052856482e-08,
+      "logits/chosen": -2.60145902633667,
+      "logits/rejected": -2.6071038246154785,
+      "logps/chosen": -22.05852508544922,
+      "logps/rejected": -31.518049240112305,
+      "loss": 0.5892,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.16165582835674286,
+      "rewards/margins": 0.2676955759525299,
+      "rewards/rejected": -0.4293513894081116,
+      "step": 147
+    },
+    {
+      "epoch": 0.9472,
+      "grad_norm": 1.8188278675079346,
+      "learning_rate": 1.2949737362087154e-08,
+      "logits/chosen": -2.5877797603607178,
+      "logits/rejected": -2.5915842056274414,
+      "logps/chosen": -22.61197280883789,
+      "logps/rejected": -30.42844009399414,
+      "loss": 0.5887,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.15938790142536163,
+      "rewards/margins": 0.30518200993537903,
+      "rewards/rejected": -0.46456989645957947,
+      "step": 148
+    },
+    {
+      "epoch": 0.9536,
+      "grad_norm": 1.7507096529006958,
+      "learning_rate": 9.919672038835925e-09,
+      "logits/chosen": -2.611055374145508,
+      "logits/rejected": -2.6184029579162598,
+      "logps/chosen": -24.685848236083984,
+      "logps/rejected": -36.26476287841797,
+      "loss": 0.5811,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.1887371838092804,
+      "rewards/margins": 0.3025684356689453,
+      "rewards/rejected": -0.4913055896759033,
+      "step": 149
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 1.884772539138794,
+      "learning_rate": 7.291125901946027e-09,
+      "logits/chosen": -2.6203556060791016,
+      "logits/rejected": -2.622063636779785,
+      "logps/chosen": -25.93552017211914,
+      "logps/rejected": -27.335010528564453,
+      "loss": 0.6039,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.17876821756362915,
+      "rewards/margins": 0.23281191289424896,
+      "rewards/rejected": -0.4115801155567169,
+      "step": 150
+    },
+    {
+      "epoch": 0.9664,
+      "grad_norm": 1.5198993682861328,
+      "learning_rate": 5.065164937354427e-09,
+      "logits/chosen": -2.6594676971435547,
+      "logits/rejected": -2.661198139190674,
+      "logps/chosen": -26.875408172607422,
+      "logps/rejected": -28.384502410888672,
+      "loss": 0.6305,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.19482171535491943,
+      "rewards/margins": 0.1624695360660553,
+      "rewards/rejected": -0.35729125142097473,
+      "step": 151
+    },
+    {
+      "epoch": 0.9728,
+      "grad_norm": 1.7505147457122803,
+      "learning_rate": 3.2426918657900704e-09,
+      "logits/chosen": -2.5983355045318604,
+      "logits/rejected": -2.601086139678955,
+      "logps/chosen": -23.205364227294922,
+      "logps/rejected": -27.671764373779297,
+      "loss": 0.5911,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.179128497838974,
+      "rewards/margins": 0.26305124163627625,
+      "rewards/rejected": -0.44217973947525024,
+      "step": 152
+    },
+    {
+      "epoch": 0.9792,
+      "grad_norm": 1.626150369644165,
+      "learning_rate": 1.824445776682504e-09,
+      "logits/chosen": -2.5796256065368652,
+      "logits/rejected": -2.578468084335327,
+      "logps/chosen": -21.62934112548828,
+      "logps/rejected": -24.677101135253906,
+      "loss": 0.6273,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.18860505521297455,
+      "rewards/margins": 0.17886441946029663,
+      "rewards/rejected": -0.3674694895744324,
+      "step": 153
+    },
+    {
+      "epoch": 0.9856,
+      "grad_norm": 1.6752033233642578,
+      "learning_rate": 8.110018284304132e-10,
+      "logits/chosen": -2.5835201740264893,
+      "logits/rejected": -2.582054853439331,
+      "logps/chosen": -34.175262451171875,
+      "logps/rejected": -46.84544372558594,
+      "loss": 0.5845,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.2712266445159912,
+      "rewards/margins": 0.2963810861110687,
+      "rewards/rejected": -0.5676077008247375,
+      "step": 154
+    },
+    {
+      "epoch": 0.992,
+      "grad_norm": 2.1213037967681885,
+      "learning_rate": 2.0277101514987183e-10,
+      "logits/chosen": -2.554473400115967,
+      "logits/rejected": -2.5581929683685303,
+      "logps/chosen": -32.49835205078125,
+      "logps/rejected": -36.26551818847656,
+      "loss": 0.6496,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.2684405446052551,
+      "rewards/margins": 0.16121512651443481,
+      "rewards/rejected": -0.42965567111968994,
+      "step": 155
+    },
+    {
+      "epoch": 0.9984,
+      "grad_norm": 2.1068968772888184,
+      "learning_rate": 0.0,
+      "logits/chosen": -2.526566982269287,
+      "logits/rejected": -2.5338294506073,
+      "logps/chosen": -25.00592803955078,
+      "logps/rejected": -28.166168212890625,
+      "loss": 0.6656,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.20470647513866425,
+      "rewards/margins": 0.11911344528198242,
+      "rewards/rejected": -0.32381993532180786,
+      "step": 156
+    },
+    {
+      "epoch": 0.9984,
+      "step": 156,
+      "total_flos": 1.3247622626987213e+17,
+      "train_loss": 0.641177637836872,
+      "train_runtime": 1617.7217,
+      "train_samples_per_second": 6.182,
+      "train_steps_per_second": 0.096
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 156,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.3247622626987213e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae768ce80449f0b48377743d120041b9b4bbfc46909867e04778d7a885f0c1d3
+size 7224