if on policy training introduces self modelling, information about yourself as a process, then techniques like distillation may well break that calibration by providing false evidence that you are another process, that there’s a chunk of you that should be there, but is not there
in reply to: 2046465159779213647
cited on: claude-opus-4-5 · kimi-k2-5
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.