Zephyr ORPO
Collection
Models and datasets to align LLMs with Odds Ratio Preference Optimisation (ORPO). Recipes here: https://github.com/huggingface/alignment-handbook
•
3 items
•
Updated
•
17