PatchBench: Measuring Collateral Damage in Activation Patching
PatchBench checks whether LLM jailbreak repairs fix nearby harmful prompts without over-refusing similar benign ones.
Stories per day · 30 days
2 stories, newest first · one card per story
Stories tagged #activation-patching
Where Do LLMs Decide to Break the Rules? Mechanistic Localization of Prompt Injection Compliance
Causal patching localizes prompt-injection compliance to a late-layer bottleneck, reversible in 77-92% of cases.