SOURCE-LINKED INTELLIGENCE

Steering Interference Reflects the Model's Defaults, Not the Behavior Directions

arXiv · AI, language, vision and robotics · article · Sep 7, 2026 · UTC

Activation steering promises modular control of language model behavior: a behavior such as politeness corresponds to a direction in a model's activations, and adding that direction while it generates should switch the behavior on and leave everything else alone. It does not. We ask what decides which other behaviors move, and by how much, and find that it is the model rather than the behavior being steered. A steer relaxes the model toward a small set of behaviors it already favors, chiefly refusal, sycophancy, and poeticism, and that set is much the same whatever is steered. Three results ac

Read original source ↗ Open in workspace

recordType

paper

region

Global