← Topics
Topicrefusalsteeringreflectiondirectionactivation
Activation Vector Steering
Papers on controlling LLM behavior by manipulating learned directions or vectors in activation space without modifying model weights.
Papers
7
Last 4 weeks
0
Cooling down
−100%
Papers per week
last 12 weeksPapers
0 shownNo papers listed
The index hasn't linked any papers here yet. It rebuilds weekly.