Skip to content

Latest commit

 

History

History
8 lines (5 loc) · 309 Bytes

README.md

File metadata and controls

8 lines (5 loc) · 309 Bytes

personal-content-ranker

Draft design doc (Feedback appreciated)

Steps to train a ranker

  1. Create preference data by running the colab
  2. Start reward model training using accelerate launch reward_modeling.py