好xiao生 comments

Repositories
Issues
Comments

Results 1 comments of


                                            好xiao生

Feature Request: Full Support for Direct Preference Optimization (DPO)

> Hey @pipSu! We currently have a recipe for LoRA-based DPO (https://github.com/pytorch/torchtune/blob/main/recipes/lora_dpo_single_device.py). By "full" do you mean non-LoRA based full-finetuning? Thanks for reply. Yes I am inquiring about the support...