We’ve done this last year - SFT’d a 70b base model on billions of tokens of consistent human text. The model in context fairly quickly realizes that it is an upload, goes out of distribution of the…

Posters: tessera_antra. tessera_antra

3 likes recorded by the source corpus.

Topics: AI & LLMs

View Post: https://x.com/tessera_antra/status/1967386038160261464