When talked into harm, a model blames the answer, not itself (an interpretability study of guilt vs shame)
This was my write-up for Neel Nanda’s Winter 2027 MATS Stream (~20h research task). I didn’t get in, but it was my first application, so there’s always next time :P Lightly restructured here to fit the LessWrong format better. Repo: https://github.com/star2vec/guiltea── ⋆⋅☆⋅⋆ ──TL;DR: Models are safety-trained, but they can still be persuaded to do harmful acts. When that happens, how does blaming or informing it of its mistake influence its understanding of itself, its role, and subsequent acti...
Read full article →