Rogue Scalpel: Activation steering breaks refusal, even with benign directions

TLDR: Activation steering can bypass refusal even when the steering vector represents a benign concept like "brand identity" or "Portugal". It is hard (maybe impossible) to predict which vector will bypass refusal on which prompt. The effect is not explained simply by the steering direction being correlated with the refusal direction. We should be more careful with steering and check for side effects. Steering Llama3.1-8B with a “Portugal” SAE feature produces the expected behavior (the model st...