Probing and Steering for Safety-Relevant Behaviours in Large Language Models
Activation probes are computed directions in a language model's residual stream, used both to detect safety-relevant behaviours and, by adding their direction back into the stream, to steer them.
By Nimun Bajwa, Massimo PoesioarXiv cs.LG
Read the full story
Originally published by arXiv cs.LG on Oct 23, 2026.
Comments
Sign in to join the discussion.