llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-s_star-0.4-20260425-111846/README.md

library_name, base_model, tags, datasets, model-index

library_name	base_model	tags	datasets	model-index
transformers	W-61/llama-3-8b-base-sft-ultrachat-8xh200	alignment-handbook new-dpo generated_from_trainer	HuggingFaceH4/ultrafeedback_binarized	name results llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-s_star-0.4-20260425-111846

llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-s_star-0.4-20260425-111846

This model is a fine-tuned version of W-61/llama-3-8b-base-sft-ultrachat-8xh200 on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:

• Loss: 0.5784
• Fcm Dpo/beta: 0.0040
• Margin Dpo/margin Mean: 88.5980
• Margin Dpo/margin Std: 143.0663
• Logps/chosen: -496.1021
• Logps/rejected: -563.8033
• Logps/ref Chosen: -287.8268
• Logps/ref Rejected: -266.9300
• Logits/chosen: -0.8692
• Logits/rejected: -0.8529

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-07
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 8
• total_train_batch_size: 128
• total_eval_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Fcm Dpo/beta	Margin Dpo/margin Mean	Margin Dpo/margin Std	Logps/chosen	Logps/rejected	Logps/ref Chosen	Logps/ref Rejected	Logits/chosen	Logits/rejected
4.8634	0.4188	200	0.5781	0.0070	50.7633	82.2691	-385.8169	-415.6835	-287.8268	-266.9300	-0.8936	-0.8767
4.6885	0.8377	400	0.5784	0.0040	88.5980	143.0663	-496.1021	-563.8033	-287.8268	-266.9300	-0.8692	-0.8529

Framework versions

• Transformers 4.51.0
• Pytorch 2.3.1+cu121
• Datasets 2.21.0
• Tokenizers 0.21.4