@zswitten the reason it's interesting to look at the behavior downstream of any 'early sandbagging' is that you get to see how 'insistent' it is, in what circumstances, etc.maybe early sandbagging…

Posters: anthrupad. anthrupad

2 likes recorded by the source corpus.

Topics: AI & LLMs, AI safety

View Post: https://x.com/anthrupad/status/1856907137425039841