method
active
method:feature-ablation-zeroing-feature-activationsFeature ablation (zeroing feature activations)
Clamping a feature's value to zero to measure its causal effect on model output.
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Authors argue features are model properties because logit effects and ablations are consistent with feature interpretations
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.
- Intervention type that sets activations to zero, used for interpretability analysis
- Stronger test of concept isolation than steering.
- Gradient-based attribution approximates ablation impact, enabling fast search for causally important features.
- Sequential zeroing out of high-contribution heads to verify their functional specialization for persona control
- Nonspecific to self.
- Justifies SCR/TPP.
- Ablates top-k PCs of character-count means vs random subspace.
- Narrow-range greater-than feature.