method
active
method:difference-in-means-steeringDifference-in-Means Steering
Steering vector from mean activation difference between contrasting datasets.
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (2)
method
- Difference-in-means steering vectorsrelated_toUsed to extract trait and style directions.
- Activation Addition SteeringextendsAdd scaled steering vector to layer activations.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Contrast with probe steering.
- Method for extracting linear directions by subtracting mean activations of contrastive groups; used to define the Assistant Axis
- Vector from mean of false representations to mean of true representations; core of mass-mean probing
- General technique of modifying activations to control model behavior.
- Priority claim establishing novelty of the Head Cor intervention approach
- Paradigm of finding the right direction in activation space (e.g., linear steering).
- Novel method that applies intervention only when the model begins a new thinking step (at the \n\n delimiter) rather than at every token
- Using interventions to guide model generation behavior, e.g., adding sentiment vectors at inference time