Model save

Browse files

Files changed (14) hide show

README.md +80 -0
added_tokens.json +24 -0
all_results.json +9 -0
config.json +30 -0
generation_config.json +7 -0
merges.txt +0 -0
model.safetensors +3 -0
special_tokens_map.json +27 -0
tokenizer.json +0 -0
tokenizer_config.json +196 -0
train_results.json +9 -0
trainer_state.json +1219 -0
training_args.bin +3 -0
vocab.json +0 -0

README.md ADDED Viewed

	@@ -0,0 +1,80 @@

+---
+license: apache-2.0
+base_model: hZzy/qwen2.5-0.5b-sft-news-IFT
+tags:
+- trl
+- expo
+- generated_from_trainer
+model-index:
+- name: qwen2.5-0.5b-expo-DPO-EXPERIMENT-0.1-5e6
+  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. -->
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/zhiyuzha-university-of-florida/huggingface/runs/aifgjuia)
+# qwen2.5-0.5b-expo-DPO-EXPERIMENT-0.1-5e6
+This model is a fine-tuned version of [hZzy/qwen2.5-0.5b-sft-news-IFT](https://huggingface.co/hZzy/qwen2.5-0.5b-sft-news-IFT) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.7556
+- Logps: -120.0429
+- Logits: -1.9737
+- Objective: 0.7840
+- Dpo Loss: 0.7840
+- Regularize: 0.7840
+- Ranking Simple: 0.5403
+- Ranking Idealized: 0.5888
+- Ranking Idealized Expo: 0.5103
+## 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: 5e-06
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 6
+- gradient_accumulation_steps: 12
+- total_train_batch_size: 288
+- total_eval_batch_size: 24
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 2
+### Training results
+| Training Loss | Epoch  | Step | Validation Loss | Logps     | Logits  | Objective | Dpo Loss | Regularize | Ranking Simple | Ranking Idealized | Ranking Idealized Expo |
+|:-------------:|:------:|:----:|:---------------:|:---------:|:-------:|:---------:|:--------:|:----------:|:--------------:|:-----------------:|:----------------------:|
+| 0.5807        | 0.2834 | 50   | 0.6822          | -101.8611 | -1.8709 | 0.7088    | 0.7088   | 0.7088     | 0.5196         | 0.5888            | 0.5103                 |
+| 0.4908        | 0.5668 | 100  | 0.6802          | -105.0768 | -1.8507 | 0.6854    | 0.6854   | 0.6854     | 0.5300         | 0.5888            | 0.5103                 |
+| 0.4191        | 0.8503 | 150  | 0.6960          | -108.5704 | -2.1205 | 0.7127    | 0.7127   | 0.7127     | 0.5403         | 0.5888            | 0.5103                 |
+| 0.2287        | 1.1337 | 200  | 0.7276          | -115.4432 | -2.0764 | 0.7403    | 0.7403   | 0.7403     | 0.5362         | 0.5888            | 0.5103                 |
+| 0.2329        | 1.4171 | 250  | 0.7454          | -118.2405 | -2.0640 | 0.7706    | 0.7706   | 0.7706     | 0.5351         | 0.5888            | 0.5103                 |
+| 0.2036        | 1.7005 | 300  | 0.7574          | -120.7682 | -1.9746 | 0.7851    | 0.7851   | 0.7851     | 0.5434         | 0.5888            | 0.5103                 |
+| 0.2102        | 1.9839 | 350  | 0.7556          | -120.0429 | -1.9737 | 0.7840    | 0.7840   | 0.7840     | 0.5403         | 0.5888            | 0.5103                 |
+### Framework versions
+- Transformers 4.42.0
+- Pytorch 2.3.0+cu121
+- Datasets 2.19.1
+- Tokenizers 0.19.1

added_tokens.json ADDED Viewed

	@@ -0,0 +1,24 @@

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

all_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 1.995276334435522,
+    "total_flos": 0.0,
+    "train_loss": 0.36798647613349283,
+    "train_runtime": 14198.7063,
+    "train_samples": 50802,
+    "train_samples_per_second": 7.156,
+    "train_steps_per_second": 0.025
+}

config.json ADDED Viewed

	@@ -0,0 +1,30 @@

+{
+  "_name_or_path": "hZzy/qwen2.5-0.5b-sft-news-IFT",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151644,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 896,
+  "initializer_range": 0.02,
+  "intermediate_size": 4864,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 24,
+  "model_type": "qwen2",
+  "num_attention_heads": 14,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151645,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.42.0",
+  "use_cache": false,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 151665
+}

generation_config.json ADDED Viewed

	@@ -0,0 +1,7 @@

+{
+  "bos_token_id": 151644,
+  "eos_token_id": 151645,
+  "max_new_tokens": 2048,
+  "pad_token_id": 151645,
+  "transformers_version": "4.42.0"
+}

merges.txt ADDED Viewed

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

model.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:46f1024eeaccdc13838e42d11bffa39c711199540901490cda4e46471a48a614
+size 1975192208

special_tokens_map.json ADDED Viewed

	@@ -0,0 +1,27 @@

+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": {
+    "content": "<|im_start|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json ADDED Viewed

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

tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,196 @@

+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": "<|im_start|>",
+  "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n'  + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 2048,
+  "pad_token": "<|im_end|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

train_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 1.995276334435522,
+    "total_flos": 0.0,
+    "train_loss": 0.36798647613349283,
+    "train_runtime": 14198.7063,
+    "train_samples": 50802,
+    "train_samples_per_second": 7.156,
+    "train_steps_per_second": 0.025
+}

trainer_state.json ADDED Viewed

	@@ -0,0 +1,1219 @@

+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.995276334435522,
+  "eval_steps": 50,
+  "global_step": 352,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "dpo_loss": 0.6931471824645996,
+      "epoch": 0.005668398677373642,
+      "grad_norm": 13.413641782833313,
+      "learning_rate": 1.3888888888888888e-07,
+      "logits": -1.3147305250167847,
+      "logps": -88.0877456665039,
+      "loss": 0.6931,
+      "objective": 0.6931471824645996,
+      "ranking_idealized": 0.625,
+      "ranking_idealized_expo": 0.5208333134651184,
+      "ranking_simple": 0.5208333134651184,
+      "regularize": 0.6931471824645996,
+      "step": 1
+    },
+    {
+      "dpo_loss": 0.6931948065757751,
+      "epoch": 0.02834199338686821,
+      "grad_norm": 13.69587680989851,
+      "learning_rate": 6.944444444444446e-07,
+      "logits": -1.3683627843856812,
+      "logps": -83.96967315673828,
+      "loss": 0.693,
+      "objective": 0.6931948065757751,
+      "ranking_idealized": 0.6145833134651184,
+      "ranking_idealized_expo": 0.546875,
+      "ranking_simple": 0.546875,
+      "regularize": 0.6931948065757751,
+      "step": 5
+    },
+    {
+      "dpo_loss": 0.6823691129684448,
+      "epoch": 0.05668398677373642,
+      "grad_norm": 12.80657019274692,
+      "learning_rate": 1.3888888888888892e-06,
+      "logits": -1.4495999813079834,
+      "logps": -81.29175567626953,
+      "loss": 0.688,
+      "objective": 0.6823691129684448,
+      "ranking_idealized": 0.6000000238418579,
+      "ranking_idealized_expo": 0.512499988079071,
+      "ranking_simple": 0.5208333134651184,
+      "regularize": 0.6823691129684448,
+      "step": 10
+    },
+    {
+      "dpo_loss": 0.6785086989402771,
+      "epoch": 0.08502598016060463,
+      "grad_norm": 12.407233192434104,
+      "learning_rate": 2.0833333333333334e-06,
+      "logits": -1.4651646614074707,
+      "logps": -80.61344909667969,
+      "loss": 0.6772,
+      "objective": 0.6785086989402771,
+      "ranking_idealized": 0.5833333134651184,
+      "ranking_idealized_expo": 0.5041666626930237,
+      "ranking_simple": 0.5166666507720947,
+      "regularize": 0.6785086989402771,
+      "step": 15
+    },
+    {
+      "dpo_loss": 0.6960199475288391,
+      "epoch": 0.11336797354747284,
+      "grad_norm": 14.590968333806934,
+      "learning_rate": 2.7777777777777783e-06,
+      "logits": -1.5296542644500732,
+      "logps": -82.23778533935547,
+      "loss": 0.6632,
+      "objective": 0.6960199475288391,
+      "ranking_idealized": 0.625,
+      "ranking_idealized_expo": 0.5375000238418579,
+      "ranking_simple": 0.5416666865348816,
+      "regularize": 0.6960199475288391,
+      "step": 20
+    },
+    {
+      "dpo_loss": 0.6467238068580627,
+      "epoch": 0.14170996693434104,
+      "grad_norm": 12.356578847477945,
+      "learning_rate": 3.4722222222222224e-06,
+      "logits": -1.6228190660476685,
+      "logps": -81.19103240966797,
+      "loss": 0.651,
+      "objective": 0.6467238068580627,
+      "ranking_idealized": 0.5958333611488342,
+      "ranking_idealized_expo": 0.4958333373069763,
+      "ranking_simple": 0.5375000238418579,
+      "regularize": 0.6467238068580627,
+      "step": 25
+    },
+    {
+      "dpo_loss": 0.6353517174720764,
+      "epoch": 0.17005196032120926,
+      "grad_norm": 11.773480191521392,
+      "learning_rate": 4.166666666666667e-06,
+      "logits": -1.6215683221817017,
+      "logps": -84.20466613769531,
+      "loss": 0.6315,
+      "objective": 0.6353517174720764,
+      "ranking_idealized": 0.6041666865348816,
+      "ranking_idealized_expo": 0.5416666865348816,
+      "ranking_simple": 0.5625,
+      "regularize": 0.6353517174720764,
+      "step": 30
+    },
+    {
+      "dpo_loss": 0.6039083003997803,
+      "epoch": 0.19839395370807747,
+      "grad_norm": 13.700920213341456,
+      "learning_rate": 4.861111111111111e-06,
+      "logits": -1.6290026903152466,
+      "logps": -88.07846069335938,
+      "loss": 0.5989,
+      "objective": 0.6039083003997803,
+      "ranking_idealized": 0.6166666746139526,
+      "ranking_idealized_expo": 0.5041666626930237,
+      "ranking_simple": 0.574999988079071,
+      "regularize": 0.6039083003997803,
+      "step": 35
+    },
+    {
+      "dpo_loss": 0.5691289901733398,
+      "epoch": 0.22673594709494568,
+      "grad_norm": 15.376644512330481,
+      "learning_rate": 4.998023493068255e-06,
+      "logits": -1.660142183303833,
+      "logps": -92.43869018554688,
+      "loss": 0.5901,
+      "objective": 0.5691289901733398,
+      "ranking_idealized": 0.6000000238418579,
+      "ranking_idealized_expo": 0.5208333134651184,
+      "ranking_simple": 0.5916666388511658,
+      "regularize": 0.5691289901733398,
+      "step": 40
+    },
+    {
+      "dpo_loss": 0.54946368932724,
+      "epoch": 0.25507794048181387,
+      "grad_norm": 13.285916184549091,
+      "learning_rate": 4.989999289644993e-06,
+      "logits": -1.8279616832733154,
+      "logps": -92.55197143554688,
+      "loss": 0.5808,
+      "objective": 0.54946368932724,
+      "ranking_idealized": 0.6208333373069763,
+      "ranking_idealized_expo": 0.5375000238418579,
+      "ranking_simple": 0.6000000238418579,
+      "regularize": 0.54946368932724,
+      "step": 45
+    },
+    {
+      "dpo_loss": 0.5661661028862,
+      "epoch": 0.2834199338686821,
+      "grad_norm": 12.64511853627035,
+      "learning_rate": 4.975823666181256e-06,
+      "logits": -1.8111599683761597,
+      "logps": -91.70741271972656,
+      "loss": 0.5807,
+      "objective": 0.5661661028862,
+      "ranking_idealized": 0.5416666865348816,
+      "ranking_idealized_expo": 0.42500001192092896,
+      "ranking_simple": 0.5833333134651184,
+      "regularize": 0.5661661028862,
+      "step": 50
+    },
+    {
+      "epoch": 0.2834199338686821,
+      "eval_dpo_loss": 0.7087637782096863,
+      "eval_logits": -1.8708840608596802,
+      "eval_logps": -101.86113739013672,
+      "eval_loss": 0.6822364926338196,
+      "eval_objective": 0.7087637782096863,
+      "eval_ranking_idealized": 0.5888429880142212,
+      "eval_ranking_idealized_expo": 0.5103305578231812,
+      "eval_ranking_simple": 0.51962810754776,
+      "eval_regularize": 0.7087637782096863,
+      "eval_runtime": 260.7452,
+      "eval_samples_per_second": 22.206,
+      "eval_steps_per_second": 0.928,
+      "step": 50
+    },
+    {
+      "dpo_loss": 0.5751214623451233,
+      "epoch": 0.3117619272555503,
+      "grad_norm": 13.121793817706566,
+      "learning_rate": 4.955531642853404e-06,
+      "logits": -1.8980281352996826,
+      "logps": -95.49773406982422,
+      "loss": 0.5587,
+      "objective": 0.5751214623451233,
+      "ranking_idealized": 0.625,
+      "ranking_idealized_expo": 0.5416666865348816,
+      "ranking_simple": 0.6000000238418579,
+      "regularize": 0.5751214623451233,
+      "step": 55
+    },
+    {
+      "dpo_loss": 0.5370620489120483,
+      "epoch": 0.3401039206424185,
+      "grad_norm": 14.066207842989309,
+      "learning_rate": 4.929173350101025e-06,
+      "logits": -1.8723183870315552,
+      "logps": -95.54426574707031,
+      "loss": 0.5636,
+      "objective": 0.5370620489120483,
+      "ranking_idealized": 0.637499988079071,
+      "ranking_idealized_expo": 0.512499988079071,
+      "ranking_simple": 0.612500011920929,
+      "regularize": 0.5370620489120483,
+      "step": 60
+    },
+    {
+      "dpo_loss": 0.518057644367218,
+      "epoch": 0.3684459140292867,
+      "grad_norm": 11.034128756662287,
+      "learning_rate": 4.896813904782162e-06,
+      "logits": -1.8221579790115356,
+      "logps": -94.84842681884766,
+      "loss": 0.5446,
+      "objective": 0.518057644367218,
+      "ranking_idealized": 0.5708333253860474,
+      "ranking_idealized_expo": 0.5083333253860474,
+      "ranking_simple": 0.6333333253860474,
+      "regularize": 0.518057644367218,
+      "step": 65
+    },
+    {
+      "dpo_loss": 0.5486761331558228,
+      "epoch": 0.39678790741615494,
+      "grad_norm": 13.022002107607332,
+      "learning_rate": 4.858533249305337e-06,
+      "logits": -1.8116071224212646,
+      "logps": -95.58805847167969,
+      "loss": 0.5527,
+      "objective": 0.5486761331558228,
+      "ranking_idealized": 0.6000000238418579,
+      "ranking_idealized_expo": 0.5166666507720947,
+      "ranking_simple": 0.6041666865348816,
+      "regularize": 0.5486761331558228,
+      "step": 70
+    },
+    {
+      "dpo_loss": 0.5020822286605835,
+      "epoch": 0.42512990080302315,
+      "grad_norm": 13.443795403497152,
+      "learning_rate": 4.814425954135786e-06,
+      "logits": -1.8384398221969604,
+      "logps": -95.35403442382812,
+      "loss": 0.506,
+      "objective": 0.5020822286605835,
+      "ranking_idealized": 0.6291666626930237,
+      "ranking_idealized_expo": 0.5458333492279053,
+      "ranking_simple": 0.6583333611488342,
+      "regularize": 0.5020822286605835,
+      "step": 75
+    },
+    {
+      "dpo_loss": 0.5126288533210754,
+      "epoch": 0.45347189418989137,
+      "grad_norm": 14.542906941369981,
+      "learning_rate": 4.764600984163809e-06,
+      "logits": -1.8958051204681396,
+      "logps": -94.8360824584961,
+      "loss": 0.5222,
+      "objective": 0.5126288533210754,
+      "ranking_idealized": 0.5833333134651184,
+      "ranking_idealized_expo": 0.5166666507720947,
+      "ranking_simple": 0.675000011920929,
+      "regularize": 0.5126288533210754,
+      "step": 80
+    },
+    {
+      "dpo_loss": 0.4901658594608307,
+      "epoch": 0.4818138875767596,
+      "grad_norm": 12.218149677924327,
+      "learning_rate": 4.709181429512391e-06,
+      "logits": -1.8605338335037231,
+      "logps": -91.9226303100586,
+      "loss": 0.5101,
+      "objective": 0.4901658594608307,
+      "ranking_idealized": 0.6000000238418579,
+      "ranking_idealized_expo": 0.5249999761581421,
+      "ranking_simple": 0.6333333253860474,
+      "regularize": 0.4901658594608307,
+      "step": 85
+    },
+    {
+      "dpo_loss": 0.5431678891181946,
+      "epoch": 0.5101558809636277,
+      "grad_norm": 12.479759342264707,
+      "learning_rate": 4.648304201449153e-06,
+      "logits": -1.6652709245681763,
+      "logps": -94.42205047607422,
+      "loss": 0.4784,
+      "objective": 0.5431678891181946,
+      "ranking_idealized": 0.550000011920929,
+      "ranking_idealized_expo": 0.4791666567325592,
+      "ranking_simple": 0.5833333134651184,
+      "regularize": 0.5431678891181946,
+      "step": 90
+    },
+    {
+      "dpo_loss": 0.45254868268966675,
+      "epoch": 0.538497874350496,
+      "grad_norm": 13.513215812074137,
+      "learning_rate": 4.582119694153834e-06,
+      "logits": -1.7359992265701294,
+      "logps": -94.93058013916016,
+      "loss": 0.4763,
+      "objective": 0.45254868268966675,
+      "ranking_idealized": 0.625,
+      "ranking_idealized_expo": 0.5416666865348816,
+      "ranking_simple": 0.6833333373069763,
+      "regularize": 0.45254868268966675,
+      "step": 95
+    },
+    {
+      "dpo_loss": 0.4402621388435364,
+      "epoch": 0.5668398677373642,
+      "grad_norm": 13.595765492893772,
+      "learning_rate": 4.510791413176912e-06,
+      "logits": -1.849007248878479,
+      "logps": -101.81654357910156,
+      "loss": 0.4908,
+      "objective": 0.4402621388435364,
+      "ranking_idealized": 0.612500011920929,
+      "ranking_idealized_expo": 0.5416666865348816,
+      "ranking_simple": 0.7041666507720947,
+      "regularize": 0.4402621388435364,
+      "step": 100
+    },
+    {
+      "epoch": 0.5668398677373642,
+      "eval_dpo_loss": 0.6854463219642639,
+      "eval_logits": -1.8507442474365234,
+      "eval_logps": -105.0768051147461,
+      "eval_loss": 0.6801603436470032,
+      "eval_objective": 0.6854463219642639,
+      "eval_ranking_idealized": 0.5888429880142212,
+      "eval_ranking_idealized_expo": 0.5103305578231812,
+      "eval_ranking_simple": 0.5299586653709412,
+      "eval_regularize": 0.6854463219642639,
+      "eval_runtime": 259.5573,
+      "eval_samples_per_second": 22.307,
+      "eval_steps_per_second": 0.932,
+      "step": 100
+    },
+    {
+      "dpo_loss": 0.43196719884872437,
+      "epoch": 0.5951818611242324,
+      "grad_norm": 11.229649770242489,
+      "learning_rate": 4.4344955715072344e-06,
+      "logits": -1.6958547830581665,
+      "logps": -102.08232879638672,
+      "loss": 0.4538,
+      "objective": 0.43196719884872437,
+      "ranking_idealized": 0.6208333373069763,
+      "ranking_idealized_expo": 0.5333333611488342,
+      "ranking_simple": 0.675000011920929,
+      "regularize": 0.43196719884872437,
+      "step": 105
+    },
+    {
+      "dpo_loss": 0.4712914526462555,
+      "epoch": 0.6235238545111006,
+      "grad_norm": 12.961407259435642,
+      "learning_rate": 4.353420654246546e-06,
+      "logits": -1.8904917240142822,
+      "logps": -100.97926330566406,
+      "loss": 0.4539,
+      "objective": 0.4712914526462555,
+      "ranking_idealized": 0.5708333253860474,
+      "ranking_idealized_expo": 0.4541666805744171,
+      "ranking_simple": 0.6666666865348816,
+      "regularize": 0.4712914526462555,
+      "step": 110
+    },
+    {
+      "dpo_loss": 0.45986562967300415,
+      "epoch": 0.6518658478979689,
+      "grad_norm": 12.814230479704358,
+      "learning_rate": 4.267766952966369e-06,
+      "logits": -1.8556350469589233,
+      "logps": -102.53894805908203,
+      "loss": 0.4603,
+      "objective": 0.45986562967300415,
+      "ranking_idealized": 0.574999988079071,
+      "ranking_idealized_expo": 0.4833333194255829,
+      "ranking_simple": 0.6208333373069763,
+      "regularize": 0.45986562967300415,
+      "step": 115
+    },
+    {
+      "dpo_loss": 0.4761102795600891,
+      "epoch": 0.680207841284837,
+      "grad_norm": 12.618691610478947,
+      "learning_rate": 4.177746070897593e-06,
+      "logits": -1.7863221168518066,
+      "logps": -104.2935791015625,
+      "loss": 0.4497,
+      "objective": 0.4761102795600891,
+      "ranking_idealized": 0.5874999761581421,
+      "ranking_idealized_expo": 0.5333333611488342,
+      "ranking_simple": 0.6583333611488342,
+      "regularize": 0.4761102795600891,
+      "step": 120
+    },
+    {
+      "dpo_loss": 0.4552411735057831,
+      "epoch": 0.7085498346717053,
+      "grad_norm": 11.61517994185661,
+      "learning_rate": 4.083580400175153e-06,
+      "logits": -1.7287298440933228,
+      "logps": -104.63223266601562,
+      "loss": 0.4684,
+      "objective": 0.4552411735057831,
+      "ranking_idealized": 0.6291666626930237,
+      "ranking_idealized_expo": 0.550000011920929,
+      "ranking_simple": 0.6958333253860474,
+      "regularize": 0.4552411735057831,
+      "step": 125
+    },
+    {
+      "dpo_loss": 0.4498961567878723,
+      "epoch": 0.7368918280585735,
+      "grad_norm": 12.014560354642498,
+      "learning_rate": 3.985502572429277e-06,
+      "logits": -1.8877291679382324,
+      "logps": -98.7514877319336,
+      "loss": 0.4427,
+      "objective": 0.4498961567878723,
+      "ranking_idealized": 0.5416666865348816,
+      "ranking_idealized_expo": 0.44999998807907104,
+      "ranking_simple": 0.6499999761581421,
+      "regularize": 0.4498961567878723,
+      "step": 130
+    },
+    {
+      "dpo_loss": 0.5134549140930176,
+      "epoch": 0.7652338214454416,
+      "grad_norm": 13.034967407252013,
+      "learning_rate": 3.8837548840805395e-06,
+      "logits": -1.9060827493667603,
+      "logps": -102.064697265625,
+      "loss": 0.4396,
+      "objective": 0.5134549140930176,
+      "ranking_idealized": 0.5833333134651184,
+      "ranking_idealized_expo": 0.5291666388511658,
+      "ranking_simple": 0.6666666865348816,
+      "regularize": 0.5134549140930176,
+      "step": 135
+    },
+    {
+      "dpo_loss": 0.4010888636112213,
+      "epoch": 0.7935758148323099,
+      "grad_norm": 11.920508064913742,
+      "learning_rate": 3.7785886977585562e-06,
+      "logits": -1.866942048072815,
+      "logps": -102.13489532470703,
+      "loss": 0.4133,
+      "objective": 0.4010888636112213,
+      "ranking_idealized": 0.5791666507720947,
+      "ranking_idealized_expo": 0.5041666626930237,
+      "ranking_simple": 0.6875,
+      "regularize": 0.4010888636112213,
+      "step": 140
+    },
+    {
+      "dpo_loss": 0.3797028958797455,
+      "epoch": 0.821917808219178,
+      "grad_norm": 13.093355264813445,
+      "learning_rate": 3.6702638213230344e-06,
+      "logits": -1.9820648431777954,
+      "logps": -102.35152435302734,
+      "loss": 0.4038,
+      "objective": 0.3797028958797455,
+      "ranking_idealized": 0.5791666507720947,
+      "ranking_idealized_expo": 0.4833333194255829,
+      "ranking_simple": 0.7250000238418579,
+      "regularize": 0.3797028958797455,
+      "step": 145
+    },
+    {
+      "dpo_loss": 0.4191714823246002,
+      "epoch": 0.8502598016060463,
+      "grad_norm": 11.55897727937339,
+      "learning_rate": 3.5590478660213214e-06,
+      "logits": -2.00236439704895,
+      "logps": -100.3611831665039,
+      "loss": 0.4191,
+      "objective": 0.4191714823246002,
+      "ranking_idealized": 0.6166666746139526,
+      "ranking_idealized_expo": 0.5416666865348816,
+      "ranking_simple": 0.7208333611488342,
+      "regularize": 0.4191714823246002,
+      "step": 150
+    },
+    {
+      "epoch": 0.8502598016060463,
+      "eval_dpo_loss": 0.7127470374107361,
+      "eval_logits": -2.1205222606658936,
+      "eval_logps": -108.57035827636719,
+      "eval_loss": 0.695988655090332,
+      "eval_objective": 0.7127470374107361,
+      "eval_ranking_idealized": 0.5888429880142212,
+      "eval_ranking_idealized_expo": 0.5103305578231812,
+      "eval_ranking_simple": 0.5402892827987671,
+      "eval_regularize": 0.7127470374107361,
+      "eval_runtime": 259.6667,
+      "eval_samples_per_second": 22.298,
+      "eval_steps_per_second": 0.932,
+      "step": 150
+    },
+    {
+      "dpo_loss": 0.394189715385437,
+      "epoch": 0.8786017949929145,
+      "grad_norm": 13.2018803660899,
+      "learning_rate": 3.4452155853680457e-06,
+      "logits": -2.0352015495300293,
+      "logps": -101.7432632446289,
+      "loss": 0.387,
+      "objective": 0.394189715385437,
+      "ranking_idealized": 0.6166666746139526,
+      "ranking_idealized_expo": 0.512499988079071,
+      "ranking_simple": 0.7166666388511658,
+      "regularize": 0.394189715385437,
+      "step": 155
+    },
+    {
+      "dpo_loss": 0.41441217064857483,
+      "epoch": 0.9069437883797827,
+      "grad_norm": 13.589086232542883,
+      "learning_rate": 3.32904819638017e-06,
+      "logits": -1.929869532585144,
+      "logps": -104.0088119506836,
+      "loss": 0.4188,
+      "objective": 0.41441217064857483,
+      "ranking_idealized": 0.5708333253860474,
+      "ranking_idealized_expo": 0.4958333373069763,
+      "ranking_simple": 0.699999988079071,
+      "regularize": 0.41441217064857483,
+      "step": 160
+    },
+    {
+      "dpo_loss": 0.4174909293651581,
+      "epoch": 0.9352857817666509,
+      "grad_norm": 12.769842629759419,
+      "learning_rate": 3.2108326848442507e-06,
+      "logits": -1.9259008169174194,
+      "logps": -102.89508819580078,
+      "loss": 0.3934,
+      "objective": 0.4174909293651581,
+      "ranking_idealized": 0.6041666865348816,
+      "ranking_idealized_expo": 0.5208333134651184,
+      "ranking_simple": 0.7083333134651184,
+      "regularize": 0.4174909293651581,
+      "step": 165
+    },
+    {
+      "dpo_loss": 0.3482740819454193,
+      "epoch": 0.9636277751535192,
+      "grad_norm": 12.654714223723412,
+      "learning_rate": 3.090861096332263e-06,
+      "logits": -1.8867944478988647,
+      "logps": -103.65254974365234,
+      "loss": 0.385,
+      "objective": 0.3482740819454193,
+      "ranking_idealized": 0.5791666507720947,
+      "ranking_idealized_expo": 0.512499988079071,
+      "ranking_simple": 0.7333333492279053,
+      "regularize": 0.3482740819454193,
+      "step": 170
+    },
+    {
+      "dpo_loss": 0.3600638806819916,
+      "epoch": 0.9919697685403873,
+      "grad_norm": 12.227920456755363,
+      "learning_rate": 2.9694298147174566e-06,
+      "logits": -1.8309239149093628,
+      "logps": -104.46839141845703,
+      "loss": 0.3687,
+      "objective": 0.3600638806819916,
+      "ranking_idealized": 0.6541666388511658,
+      "ranking_idealized_expo": 0.574999988079071,
+      "ranking_simple": 0.7833333611488342,
+      "regularize": 0.3600638806819916,
+      "step": 175
+    },
+    {
+      "dpo_loss": 0.2651214897632599,
+      "epoch": 1.0203117619272555,
+      "grad_norm": 9.967134017735377,
+      "learning_rate": 2.8468388299726714e-06,
+      "logits": -2.0111324787139893,
+      "logps": -106.2859115600586,
+      "loss": 0.2698,
+      "objective": 0.2651214897632599,
+      "ranking_idealized": 0.6166666746139526,
+      "ranking_idealized_expo": 0.5416666865348816,
+      "ranking_simple": 0.8083333373069763,
+      "regularize": 0.2651214897632599,
+      "step": 180
+    },
+    {
+      "dpo_loss": 0.2080942541360855,
+      "epoch": 1.0486537553141237,
+      "grad_norm": 13.966930247511787,
+      "learning_rate": 2.723390997059943e-06,
+      "logits": -1.873337984085083,
+      "logps": -111.36536407470703,
+      "loss": 0.2277,
+      "objective": 0.2080942541360855,
+      "ranking_idealized": 0.5791666507720947,
+      "ranking_idealized_expo": 0.4833333194255829,
+      "ranking_simple": 0.8083333373069763,
+      "regularize": 0.2080942541360855,
+      "step": 185
+    },
+    {
+      "dpo_loss": 0.23763832449913025,
+      "epoch": 1.076995748700992,
+      "grad_norm": 14.992011192414951,
+      "learning_rate": 2.599391287742315e-06,
+      "logits": -1.8715498447418213,
+      "logps": -118.10340118408203,
+      "loss": 0.2298,
+      "objective": 0.23763832449913025,
+      "ranking_idealized": 0.625,
+      "ranking_idealized_expo": 0.5541666746139526,
+      "ranking_simple": 0.8500000238418579,
+      "regularize": 0.23763832449913025,
+      "step": 190
+    },
+    {
+      "dpo_loss": 0.21301580965518951,
+      "epoch": 1.10533774208786,
+      "grad_norm": 10.744062848508776,
+      "learning_rate": 2.4751460371661763e-06,
+      "logits": -1.8871595859527588,
+      "logps": -113.93363952636719,
+      "loss": 0.2267,
+      "objective": 0.21301580965518951,
+      "ranking_idealized": 0.6333333253860474,
+      "ranking_idealized_expo": 0.5666666626930237,
+      "ranking_simple": 0.8333333134651184,
+      "regularize": 0.21301580965518951,
+      "step": 195
+    },
+    {
+      "dpo_loss": 0.21621347963809967,
+      "epoch": 1.1336797354747283,
+      "grad_norm": 9.639805136359016,
+      "learning_rate": 2.3509621870754505e-06,
+      "logits": -2.02681303024292,
+      "logps": -110.0398941040039,
+      "loss": 0.2287,
+      "objective": 0.21621347963809967,
+      "ranking_idealized": 0.6000000238418579,
+      "ranking_idealized_expo": 0.5083333253860474,
+      "ranking_simple": 0.8583333492279053,
+      "regularize": 0.21621347963809967,
+      "step": 200
+    },
+    {
+      "epoch": 1.1336797354747283,
+      "eval_dpo_loss": 0.7402901649475098,
+      "eval_logits": -2.076362133026123,
+      "eval_logps": -115.44319915771484,
+      "eval_loss": 0.7275605797767639,
+      "eval_objective": 0.7402901649475098,
+      "eval_ranking_idealized": 0.5888429880142212,
+      "eval_ranking_idealized_expo": 0.5103305578231812,
+      "eval_ranking_simple": 0.5361570119857788,
+      "eval_regularize": 0.7402901649475098,
+      "eval_runtime": 259.9944,
+      "eval_samples_per_second": 22.27,
+      "eval_steps_per_second": 0.931,
+      "step": 200
+    },
+    {
+      "dpo_loss": 0.23257608711719513,
+      "epoch": 1.1620217288615966,
+      "grad_norm": 10.185272365862717,
+      "learning_rate": 2.2271465275271985e-06,
+      "logits": -2.0131571292877197,
+      "logps": -111.7054214477539,
+      "loss": 0.2278,
+      "objective": 0.23257608711719513,
+      "ranking_idealized": 0.5708333253860474,
+      "ranking_idealized_expo": 0.48750001192092896,
+      "ranking_simple": 0.8208333253860474,
+      "regularize": 0.23257608711719513,
+      "step": 205
+    },
+    {
+      "dpo_loss": 0.2272442728281021,
+      "epoch": 1.1903637222484649,
+      "grad_norm": 9.529035858141413,
+      "learning_rate": 2.1040049389819628e-06,
+      "logits": -1.9997531175613403,
+      "logps": -110.997314453125,
+      "loss": 0.2366,
+      "objective": 0.2272442728281021,
+      "ranking_idealized": 0.5874999761581421,
+      "ranking_idealized_expo": 0.5,
+      "ranking_simple": 0.7958333492279053,
+      "regularize": 0.2272442728281021,
+      "step": 210
+    },
+    {
+      "dpo_loss": 0.2654048204421997,
+      "epoch": 1.2187057156353331,
+      "grad_norm": 11.812205063296917,
+      "learning_rate": 1.9818416366412277e-06,
+      "logits": -1.978461742401123,
+      "logps": -113.06928253173828,
+      "loss": 0.2372,
+      "objective": 0.2654048204421997,
+      "ranking_idealized": 0.5625,
+      "ranking_idealized_expo": 0.4833333194255829,
+      "ranking_simple": 0.800000011920929,
+      "regularize": 0.2654048204421997,
+      "step": 215
+    },
+    {
+      "dpo_loss": 0.2245999276638031,
+      "epoch": 1.2470477090222012,
+      "grad_norm": 12.627574377824557,
+      "learning_rate": 1.8609584188988135e-06,
+      "logits": -1.9322872161865234,
+      "logps": -112.9950942993164,
+      "loss": 0.2418,
+      "objective": 0.2245999276638031,
+      "ranking_idealized": 0.6541666388511658,
+      "ranking_idealized_expo": 0.6041666865348816,
+      "ranking_simple": 0.8583333492279053,
+      "regularize": 0.2245999276638031,
+      "step": 220
+    },
+    {
+      "dpo_loss": 0.21560445427894592,
+      "epoch": 1.2753897024090695,
+      "grad_norm": 10.156285420255232,
+      "learning_rate": 1.7416539217628792e-06,
+      "logits": -1.947004795074463,
+      "logps": -113.21937561035156,
+      "loss": 0.2349,
+      "objective": 0.21560445427894592,
+      "ranking_idealized": 0.612500011920929,
+      "ranking_idealized_expo": 0.5291666388511658,
+      "ranking_simple": 0.8416666388511658,
+      "regularize": 0.21560445427894592,
+      "step": 225
+    },
+    {
+      "dpo_loss": 0.25168582797050476,
+      "epoch": 1.3037316957959377,
+      "grad_norm": 9.19384725201928,
+      "learning_rate": 1.6242228810904393e-06,
+      "logits": -1.8978828191757202,
+      "logps": -114.50511932373047,
+      "loss": 0.2522,
+      "objective": 0.25168582797050476,
+      "ranking_idealized": 0.5833333134651184,
+      "ranking_idealized_expo": 0.5083333253860474,
+      "ranking_simple": 0.7958333492279053,
+      "regularize": 0.25168582797050476,
+      "step": 230
+    },
+    {
+      "dpo_loss": 0.20120379328727722,
+      "epoch": 1.3320736891828058,
+      "grad_norm": 9.731362112902348,
+      "learning_rate": 1.508955404457015e-06,
+      "logits": -2.018658399581909,
+      "logps": -108.0621109008789,
+      "loss": 0.2386,
+      "objective": 0.20120379328727722,
+      "ranking_idealized": 0.6041666865348816,
+      "ranking_idealized_expo": 0.4833333194255829,
+      "ranking_simple": 0.8583333492279053,
+      "regularize": 0.20120379328727722,
+      "step": 235
+    },
+    {
+      "dpo_loss": 0.2090713083744049,
+      "epoch": 1.360415682569674,
+      "grad_norm": 11.056486327313111,
+      "learning_rate": 1.3961362544602215e-06,
+      "logits": -1.977256417274475,
+      "logps": -111.13236236572266,
+      "loss": 0.2297,
+      "objective": 0.2090713083744049,
+      "ranking_idealized": 0.5541666746139526,
+      "ranking_idealized_expo": 0.4791666567325592,
+      "ranking_simple": 0.8291666507720947,
+      "regularize": 0.2090713083744049,
+      "step": 240
+    },
+    {
+      "dpo_loss": 0.2641766667366028,
+      "epoch": 1.3887576759565423,
+      "grad_norm": 10.388090633212887,
+      "learning_rate": 1.2860441452278578e-06,
+      "logits": -2.022833824157715,
+      "logps": -111.30134582519531,
+      "loss": 0.2375,
+      "objective": 0.2641766667366028,
+      "ranking_idealized": 0.5916666388511658,
+      "ranking_idealized_expo": 0.5208333134651184,
+      "ranking_simple": 0.824999988079071,
+      "regularize": 0.2641766667366028,
+      "step": 245
+    },
+    {
+      "dpo_loss": 0.21442964673042297,
+      "epoch": 1.4170996693434104,
+      "grad_norm": 12.097077417741462,
+      "learning_rate": 1.1789510538684524e-06,
+      "logits": -1.969295620918274,
+      "logps": -112.7668228149414,
+      "loss": 0.2329,
+      "objective": 0.21442964673042297,
+      "ranking_idealized": 0.574999988079071,
+      "ranking_idealized_expo": 0.47083333134651184,
+      "ranking_simple": 0.8374999761581421,
+      "regularize": 0.21442964673042297,
+      "step": 250
+    },
+    {
+      "epoch": 1.4170996693434104,
+      "eval_dpo_loss": 0.7705549597740173,
+      "eval_logits": -2.064002513885498,
+      "eval_logps": -118.24050903320312,
+      "eval_loss": 0.7454098463058472,
+      "eval_objective": 0.7705549597740173,
+      "eval_ranking_idealized": 0.5888429880142212,
+      "eval_ranking_idealized_expo": 0.5103305578231812,
+      "eval_ranking_simple": 0.5351239442825317,
+      "eval_regularize": 0.7705549597740173,
+      "eval_runtime": 259.7824,
+      "eval_samples_per_second": 22.288,
+      "eval_steps_per_second": 0.932,
+      "step": 250
+    },
+    {
+      "dpo_loss": 0.21537743508815765,
+      "epoch": 1.4454416627302786,
+      "grad_norm": 11.517471828255623,
+      "learning_rate": 1.0751215485652644e-06,
+      "logits": -1.9144450426101685,
+      "logps": -114.43687438964844,
+      "loss": 0.2261,
+      "objective": 0.21537743508815765,
+      "ranking_idealized": 0.6166666746139526,
+      "ranking_idealized_expo": 0.5458333492279053,
+      "ranking_simple": 0.8458333611488342,
+      "regularize": 0.21537743508815765,
+      "step": 255
+    },
+    {
+      "dpo_loss": 0.21364690363407135,
+      "epoch": 1.473783656117147,
+      "grad_norm": 10.723091996163696,
+      "learning_rate": 9.74812134973689e-07,
+      "logits": -1.9727046489715576,
+      "logps": -113.64453887939453,
+      "loss": 0.2051,
+      "objective": 0.21364690363407135,
+      "ranking_idealized": 0.5666666626930237,
+      "ranking_idealized_expo": 0.48750001192092896,
+      "ranking_simple": 0.8125,
+      "regularize": 0.21364690363407135,
+      "step": 260
+    },
+    {
+      "dpo_loss": 0.2255079597234726,
+      "epoch": 1.5021256495040152,
+      "grad_norm": 11.69881544405136,
+      "learning_rate": 8.78270622536716e-07,
+      "logits": -1.9215840101242065,
+      "logps": -114.519287109375,
+      "loss": 0.2204,
+      "objective": 0.2255079597234726,
+      "ranking_idealized": 0.574999988079071,
+      "ranking_idealized_expo": 0.5041666626930237,
+      "ranking_simple": 0.8208333253860474,
+      "regularize": 0.2255079597234726,
+      "step": 265
+    },
+    {
+      "dpo_loss": 0.25144073367118835,
+      "epoch": 1.5304676428908834,
+      "grad_norm": 12.543731702485788,
+      "learning_rate": 7.857355122839674e-07,
+      "logits": -1.909374713897705,
+      "logps": -116.20896911621094,
+      "loss": 0.2129,
+      "objective": 0.25144073367118835,
+      "ranking_idealized": 0.5791666507720947,
+      "ranking_idealized_expo": 0.5083333253860474,
+      "ranking_simple": 0.8083333373069763,
+      "regularize": 0.25144073367118835,
+      "step": 270
+    },
+    {
+      "dpo_loss": 0.20962868630886078,
+      "epoch": 1.5588096362777515,
+      "grad_norm": 9.738954761513112,
+      "learning_rate": 6.974354076267081e-07,
+      "logits": -1.9043574333190918,
+      "logps": -116.87390899658203,
+      "loss": 0.2209,
+      "objective": 0.20962868630886078,
+      "ranking_idealized": 0.6208333373069763,
+      "ranking_idealized_expo": 0.5249999761581421,
+      "ranking_simple": 0.8500000238418579,
+      "regularize": 0.20962868630886078,
+      "step": 275
+    },
+    {
+      "dpo_loss": 0.21624349057674408,
+      "epoch": 1.5871516296646198,
+      "grad_norm": 11.162245217664504,
+      "learning_rate": 6.135884496044245e-07,
+      "logits": -1.8271541595458984,
+      "logps": -115.82463073730469,
+      "loss": 0.2254,
+      "objective": 0.21624349057674408,
+      "ranking_idealized": 0.574999988079071,
+      "ranking_idealized_expo": 0.5041666626930237,
+      "ranking_simple": 0.8291666507720947,
+      "regularize": 0.21624349057674408,
+      "step": 280
+    },
+    {
+      "dpo_loss": 0.1944807916879654,
+      "epoch": 1.615493623051488,
+      "grad_norm": 10.89680213411274,
+      "learning_rate": 5.344017779781835e-07,
+      "logits": -1.915182113647461,
+      "logps": -117.32617950439453,
+      "loss": 0.2038,
+      "objective": 0.1944807916879654,
+      "ranking_idealized": 0.5874999761581421,
+      "ranking_idealized_expo": 0.5166666507720947,
+      "ranking_simple": 0.8416666388511658,
+      "regularize": 0.1944807916879654,
+      "step": 285
+    },
+    {
+      "dpo_loss": 0.2371363341808319,
+      "epoch": 1.643835616438356,
+      "grad_norm": 12.787089862396805,
+      "learning_rate": 4.6007101950209827e-07,
+      "logits": -1.949597716331482,
+      "logps": -114.35741424560547,
+      "loss": 0.2281,
+      "objective": 0.2371363341808319,
+      "ranking_idealized": 0.612500011920929,
+      "ranking_idealized_expo": 0.5625,
+      "ranking_simple": 0.8291666507720947,
+      "regularize": 0.2371363341808319,
+      "step": 290
+    },
+    {
+      "dpo_loss": 0.1856808066368103,
+      "epoch": 1.6721776098252243,
+      "grad_norm": 12.381384766536955,
+      "learning_rate": 3.907798046371139e-07,
+      "logits": -1.8460139036178589,
+      "logps": -113.36736297607422,
+      "loss": 0.2197,
+      "objective": 0.1856808066368103,
+      "ranking_idealized": 0.5791666507720947,
+      "ranking_idealized_expo": 0.4791666567325592,
+      "ranking_simple": 0.862500011920929,
+      "regularize": 0.1856808066368103,
+      "step": 295
+    },
+    {
+      "dpo_loss": 0.18572762608528137,
+      "epoch": 1.7005196032120926,
+      "grad_norm": 11.1669814272154,
+      "learning_rate": 3.266993139010438e-07,
+      "logits": -1.8981417417526245,
+      "logps": -116.58564758300781,
+      "loss": 0.2036,
+      "objective": 0.18572762608528137,
+      "ranking_idealized": 0.6583333611488342,
+      "ranking_idealized_expo": 0.574999988079071,
+      "ranking_simple": 0.8374999761581421,
+      "regularize": 0.18572762608528137,
+      "step": 300
+    },
+    {
+      "epoch": 1.7005196032120926,
+      "eval_dpo_loss": 0.7850732803344727,
+      "eval_logits": -1.9745572805404663,
+      "eval_logps": -120.76824188232422,
+      "eval_loss": 0.757438063621521,
+      "eval_objective": 0.7850732803344727,
+      "eval_ranking_idealized": 0.5888429880142212,
+      "eval_ranking_idealized_expo": 0.5103305578231812,
+      "eval_ranking_simple": 0.5433884263038635,
+      "eval_regularize": 0.7850732803344727,
+      "eval_runtime": 259.9996,
+      "eval_samples_per_second": 22.269,
+      "eval_steps_per_second": 0.931,
+      "step": 300
+    },
+    {
+      "dpo_loss": 0.2228098213672638,
+      "epoch": 1.7288615965989607,
+      "grad_norm": 12.534520545773153,
+      "learning_rate": 2.6798785497556184e-07,
+      "logits": -1.8615758419036865,
+      "logps": -115.37549591064453,
+      "loss": 0.2165,
+      "objective": 0.2228098213672638,
+      "ranking_idealized": 0.5625,
+      "ranking_idealized_expo": 0.5083333253860474,
+      "ranking_simple": 0.8083333373069763,
+      "regularize": 0.2228098213672638,
+      "step": 305
+    },
+    {
+      "dpo_loss": 0.20340654253959656,
+      "epoch": 1.7572035899858292,
+      "grad_norm": 11.747155200770623,
+      "learning_rate": 2.1479047161491351e-07,
+      "logits": -1.9263988733291626,
+      "logps": -112.0289535522461,
+      "loss": 0.2213,
+      "objective": 0.20340654253959656,
+      "ranking_idealized": 0.6083333492279053,
+      "ranking_idealized_expo": 0.5249999761581421,
+      "ranking_simple": 0.8500000238418579,
+      "regularize": 0.20340654253959656,
+      "step": 310
+    },
+    {
+      "dpo_loss": 0.2501530349254608,
+      "epoch": 1.7855455833726972,
+      "grad_norm": 11.283704550164217,
+      "learning_rate": 1.6723858532249783e-07,
+      "logits": -1.8143895864486694,
+      "logps": -112.80689239501953,
+      "loss": 0.2236,
+      "objective": 0.2501530349254608,
+      "ranking_idealized": 0.6041666865348816,
+      "ranking_idealized_expo": 0.5083333253860474,
+      "ranking_simple": 0.8125,
+      "regularize": 0.2501530349254608,
+      "step": 315
+    },
+    {
+      "dpo_loss": 0.21945881843566895,
+      "epoch": 1.8138875767595655,
+      "grad_norm": 11.71850155399097,
+      "learning_rate": 1.2544967068054332e-07,
+      "logits": -1.8763319253921509,
+      "logps": -117.6617202758789,
+      "loss": 0.217,
+      "objective": 0.21945881843566895,
+      "ranking_idealized": 0.6000000238418579,
+      "ranking_idealized_expo": 0.5041666626930237,
+      "ranking_simple": 0.8208333253860474,
+      "regularize": 0.21945881843566895,
+      "step": 320
+    },
+    {
+      "dpo_loss": 0.22155949473381042,
+      "epoch": 1.8422295701464337,
+      "grad_norm": 14.77256678289896,
+      "learning_rate": 8.952696513496756e-08,
+      "logits": -1.9358726739883423,
+      "logps": -119.23826599121094,
+      "loss": 0.2222,
+      "objective": 0.22155949473381042,
+      "ranking_idealized": 0.6333333253860474,
+      "ranking_idealized_expo": 0.5166666507720947,
+      "ranking_simple": 0.8666666746139526,
+      "regularize": 0.22155949473381042,
+      "step": 325
+    },
+    {
+      "dpo_loss": 0.18552295863628387,
+      "epoch": 1.8705715635333018,
+      "grad_norm": 10.882571762322252,
+      "learning_rate": 5.955921395237318e-08,
+      "logits": -1.6932176351547241,
+      "logps": -112.55815124511719,
+      "loss": 0.2239,
+      "objective": 0.18552295863628387,
+      "ranking_idealized": 0.5708333253860474,
+      "ranking_idealized_expo": 0.4749999940395355,
+      "ranking_simple": 0.8333333134651184,
+      "regularize": 0.18552295863628387,
+      "step": 330
+    },
+    {
+      "dpo_loss": 0.23661907017230988,
+      "epoch": 1.89891355692017,
+      "grad_norm": 10.436309183414208,
+      "learning_rate": 3.5620450979259034e-08,
+      "logits": -1.8299282789230347,
+      "logps": -112.96961212158203,
+      "loss": 0.2253,
+      "objective": 0.23661907017230988,
+      "ranking_idealized": 0.5916666388511658,
+      "ranking_idealized_expo": 0.4958333373069763,
+      "ranking_simple": 0.8208333253860474,
+      "regularize": 0.23661907017230988,
+      "step": 335
+    },
+    {
+      "dpo_loss": 0.20749402046203613,
+      "epoch": 1.9272555503070383,
+      "grad_norm": 10.928827594051384,
+      "learning_rate": 1.7769815745066476e-08,
+      "logits": -1.8688191175460815,
+      "logps": -116.0504379272461,
+      "loss": 0.2215,
+      "objective": 0.20749402046203613,
+      "ranking_idealized": 0.6416666507720947,
+      "ranking_idealized_expo": 0.5416666865348816,
+      "ranking_simple": 0.862500011920929,
+      "regularize": 0.20749402046203613,
+      "step": 340
+    },
+    {
+      "dpo_loss": 0.1883564591407776,
+      "epoch": 1.9555975436939064,
+      "grad_norm": 11.169724195510748,
+      "learning_rate": 6.051407360895822e-09,
+      "logits": -1.8896888494491577,
+      "logps": -116.99043273925781,
+      "loss": 0.2046,
+      "objective": 0.1883564591407776,
+      "ranking_idealized": 0.574999988079071,
+      "ranking_idealized_expo": 0.5333333611488342,
+      "ranking_simple": 0.8666666746139526,
+      "regularize": 0.1883564591407776,
+      "step": 345
+    },
+    {
+      "dpo_loss": 0.22660525143146515,
+      "epoch": 1.9839395370807746,
+      "grad_norm": 11.129465757937865,
+      "learning_rate": 4.941755748361088e-10,
+      "logits": -1.7810313701629639,
+      "logps": -115.8853530883789,
+      "loss": 0.2102,
+      "objective": 0.22660525143146515,
+      "ranking_idealized": 0.574999988079071,
+      "ranking_idealized_expo": 0.48750001192092896,
+      "ranking_simple": 0.8291666507720947,
+      "regularize": 0.22660525143146515,
+      "step": 350
+    },
+    {
+      "epoch": 1.9839395370807746,
+      "eval_dpo_loss": 0.7840016484260559,
+      "eval_logits": -1.973698377609253,
+      "eval_logps": -120.04285430908203,
+      "eval_loss": 0.7555912137031555,
+      "eval_objective": 0.7840016484260559,
+      "eval_ranking_idealized": 0.5888429880142212,
+      "eval_ranking_idealized_expo": 0.5103305578231812,
+      "eval_ranking_simple": 0.5402892827987671,
+      "eval_regularize": 0.7840016484260559,
+      "eval_runtime": 261.7375,
+      "eval_samples_per_second": 22.121,
+      "eval_steps_per_second": 0.925,
+      "step": 350
+    },
+    {
+      "epoch": 1.995276334435522,
+      "step": 352,
+      "total_flos": 0.0,
+      "train_loss": 0.36798647613349283,
+      "train_runtime": 14198.7063,
+      "train_samples_per_second": 7.156,
+      "train_steps_per_second": 0.025
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 352,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:fac51d9188e6911beb97a66bbe355150979c95ad9757ac9b032fce010c8a1d9a
+size 8184

vocab.json ADDED Viewed

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