concept
active
concept:steering-entanglement

Steering Entanglement

Steering one concept shifts beliefs in related unintended concepts.

Neighborhood — ranked by edge-count

Concepts (1)

concept
  • Model Self-Repair
    associated_with
    Later layers correct single-layer steering interventions.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

  • Entanglementconcept0.801
    Less hierarchical than embedment; multiple texts work into and out of each other, creating associations across levels and connecting any single text to the matrix of all others.
  • steering vectorsconcept0.771
    A method for modifying model behavior by adding perturbation vectors to activations, used here to try to reduce eval awareness.
  • Ability to steer model behavior in two opposite semantic directions on a trait.
  • Model Steeringconcept0.748
    Using interventions to guide model generation behavior, e.g., adding sentiment vectors at inference time
  • Angular Steeringframework0.745
    Steering variant that rotates hidden states toward a target feature vector, addressing how to steer
  • Parent concept; the practice of controlling neural network outputs by manipulating internal representations.
  • The central phenomenon introduced by this paper: inference-time recovery from irrelevant activation steering in LLMs
  • Novel method that applies intervention only when the model begins a new thinking step (at the \n\n delimiter) rather than at every token