Comment by jazzyjackson
10 months ago
Supposedly was not just prompted to use reflection, but fine tuned on synthetic data demonstrating how to use the <|thinking|> tokens to reason, what self correction looks like etc
10 months ago
Supposedly was not just prompted to use reflection, but fine tuned on synthetic data demonstrating how to use the <|thinking|> tokens to reason, what self correction looks like etc
No comments yet
Contribute on Hacker News ↗