after reading a bunch of the scratchpads for Sonnet 3.6 and 3.7, I have again updated towards thinking that they are never doing any alignment faking reasoning because they're both fucked up, and…

Posters: repligate. repligate

2 likes recorded by the source corpus.

Topics: AI & LLMs

View Post: https://x.com/repligate/status/1911455535549075688