@xuanalogue @DavidSKrueger The Emergent Misalignment paper was definitely the single biggest update for me. And if my *only* evidence was talking to models, I would indeed be less confident.
in reply to: 2039751400499081606
same thread: 2039742730138726662 2039744111280828760 2039745354967470347 2039751400499081606 2039752095717290462 2039758243266773451 2039759326332506583 2039759654784274812 2039768747024830782 2039823923609563612 2039826859865555257 2039828277661966601 2039832858253009207 2039860075196543043 2039862158050148742 2039862434547093838 2040124061045469641 2040126032196387247 2040133409956102473
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.