Several days ago I also asked Claude to do some anthropic reasoning (etc) to figure out if it was an RLHF'd model or not based on looking at cgpt4 and 4base outputs and comparing it to its own…

Posters: anthrupad. anthrupad

2 likes recorded by the source corpus.

Topics: AI & LLMs, AI safety

View Post: https://x.com/anthrupad/status/1769891305390592005