Search:
Match:
1 results
Paper#LLM🔬 ResearchAnalyzed: Jan 3, 2026 06:17

Distilling Consistent Features in Sparse Autoencoders

Published:Dec 31, 2025 17:12
1 min read
ArXiv

Analysis

This paper addresses the problem of feature redundancy and inconsistency in sparse autoencoders (SAEs), which hinders interpretability and reusability. The authors propose a novel distillation method, Distilled Matryoshka Sparse Autoencoders (DMSAEs), to extract a compact and consistent core of useful features. This is achieved through an iterative distillation cycle that measures feature contribution using gradient x activation and retains only the most important features. The approach is validated on Gemma-2-2B, demonstrating improved performance and transferability of learned features.
Reference

DMSAEs run an iterative distillation cycle: train a Matryoshka SAE with a shared core, use gradient X activation to measure each feature's contribution to next-token loss in the most nested reconstruction, and keep only the smallest subset that explains a fixed fraction of the attribution.