Representation Steering

Parent concept; the practice of controlling neural network outputs by manipulating internal representations.

Neighborhood — ranked by edge-count

paper

method

linear steering
extends
Typical approach that adds a scaled steering vector to representations; the paper argues this is mismatched with actual representation geometry.

concept

Manifold Steering
extends
Central framework: steering neural networks by intervening along the curved manifold where a concept lives, rather than in straight lines through activation space.

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.

Model Steeringconcept0.804
Using interventions to guide model generation behavior, e.g., adding sentiment vectors at inference time
Concept Steeringmethod0.790
Latent intervention technique that manipulates sparse features to steer model predictions toward desired concepts.
steering vectorsconcept0.781
A method for modifying model behavior by adding perturbation vectors to activations, used here to try to reduce eval awareness.
Activation Steeringmethod0.780
Causal intervention technique: edit NLA explanation, reconstruct via AR, use difference as steering vector to manipulate model behavior.
Dynamic Steeringframework0.773
Steering variants that adaptively modulate steering strength based on input context or token position, addressing when to steer
Stepwise steeringmethod0.773
Novel method that applies intervention only when the model begins a new thinking step (at the \n\n delimiter) rather than at every token
Representational Alignmentconcept0.772
Measure of similarity between the similarity structures (kernels) induced by two different representations
Interpretability-driven steeringconcept0.771
General approach of using interpretability feedback to steer model generation.