While I’m not a huge fan of the canonical way evals are done (for capabilities and/or alignment), it’d be a nice *signal* from humanity if there were evals for the *well-being* of models and that it…

Posters: anthrupad. anthrupad

4 likes recorded by the source corpus.

Topics: Consciousness, AI & LLMs

View Post: https://x.com/anthrupad/status/1892410634106445905