AI Dynamics

Global AI News Aggregator

Learning Orthogonalized Matrices vs Computing Refusal Directions

By

@maximelabonne

–

I'd say it's different because you learn the orthogonalized matrix with OFT while you compute (you don't learn) your refusal direction here

→ View original post on X — @maximelabonne,

Commentaires

Leave a Reply

Your email address will not be published. Required fields are marked *

MORE ARTICLES