Safety At One Shot Lm Repair

Recover safety alignment in fine-tuned LLMs using only a single safety example, without sacrificing utility. Leverages low-rank structure of safety gradients to enable minimal-cost correction even when models have been compromised by extensive harmful training data.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit 017a3bd997

Frequently asked questions

npx get-skill add adu2021/safety-at-one-shot-lm-repair