AI RESEARCH

On the Non-Identifiability of Steering Vectors in Large Language Models

arXiv CS.AI

ArXi:2602.06801v4 Announce Type: replace-cross Activation steering methods are widely used to control large language model (LLM) behavior and are often interpreted as revealing meaningful internal representations. This interpretation assumes that steering directions are identifiable and uniquely recoverable from input-output behavior. We show that, under white-box single-layer access, steering vectors are fundamentally non-identifiable due to large equivalence classes of behaviorally indistinguishable interventions.