method
active
method:l-importance-minimality-lossL_importance-minimality loss
Penalizes causal importance values above 0 with p-norm; works with p>1.
Neighborhood — ranked by edge-count
Papers (1)
paper
- Stochastic Parameter Decompositionintroducesmentions
Frameworks (1)
framework
- Linear parameter decomposition with stochastic ablation.
Claims (1)
claim
- Contrast with SDL.
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.
- Loss on randomly masked subcomponents matching target outputs, with layerwise auxiliary.
- Regularization component of the composite loss that penalizes deviation from baseline model behavior on Alpaca instructions
- Distance-based loss comparing injected representation to class centroids in the active subspace
- Auxiliary training objective from Grant (2025) that constrains intervened representations to remain near natural distribution
- Loss function pulling representations toward positive centroid and pushing away from negative centroid with angular margins
- Prior state-of-the-art baseline using the abliterated model as proposal distribution.
- A measure of whether a subcomponent is necessary to reproduce model behavior on a specific prompt, predicted by the causal importance network.
- Small per-subcomponent MLP with hard sigmoid predicting causal importance.