Pre-selection off-support ℓ1 and ℓ1/ℓ2 regularizers improve Top-k SAE monosemanticity and concentration on vision foundation models at no reconstruction cost.
arXiv:2411.02124
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 2years
2026 2roles
background 1polarities
background 1representative citing papers
Adaptive elastic net SAEs (AEN-SAEs) mitigate feature starvation in SAEs by combining ℓ2 structural stability with adaptive ℓ1 reweighting, producing a Lipschitz-continuous sparse coding map that recovers global feature support under mild assumptions.
citing papers explorer
-
Beyond the Hard Budget: Sparsity Regularizers for More Interpretable Top-k Sparse Autoencoders
Pre-selection off-support ℓ1 and ℓ1/ℓ2 regularizers improve Top-k SAE monosemanticity and concentration on vision foundation models at no reconstruction cost.
-
Feature Starvation as Geometric Instability in Sparse Autoencoders
Adaptive elastic net SAEs (AEN-SAEs) mitigate feature starvation in SAEs by combining ℓ2 structural stability with adaptive ℓ1 reweighting, producing a Lipschitz-continuous sparse coding map that recovers global feature support under mild assumptions.