cs.LGJan 30, 2026

What Does a Sparse Autoencoder Feature Do? A Weight-Based Account

Authors: Yiting Liu, Zhi-Hong Deng

Organizations: State Key Laboratory of General Artificial Intelligence, School of Intelligence Science and Technology, Peking University.

Abstract

Sparse autoencoders (SAEs) have emerged as a powerful technique for decomposing language model representations into interpretable features. Current interpretation pipelines infer feature semantics from activation patterns, implicitly assuming every feature has a semantic explanation while overlooking the computational roles features inherit from the SAE training objective. We introduce a weight-based interpretation framework that requires no activation data and grounds claims causally through targeted feature-ablation interventions. Across Gemma-2 and Llama-3.1, three depth-dependent signatures show that SAE features inherit the base model's geometry: semantic features form a U-shape under tied embeddings but bifurcate when untied, attention participation peaks mid-layer, and the two roles couple oppositely for input- vs. output-oriented features. This weight-based view supplies the causal half missing from activation-based interpretability.

Figures & tables

Appendix figures & tables35 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Are Single-Token Sparse Autoencoder Features Causally Necessary? Layer-Depth and SAE-Family Effects

    Jul 22, 2026Seonglae Cho, Zekun Wu, Kleyton Da Costa +3LLM InterpretabilitySparse Autoencoders

  2. Sparse Autoencoders Encode Both Concepts and Functions: The Downstream Geometry of Feature Effects

    Jul 27, 2026Phu Gia Hoang, Anwoy Chatterjee, Tanmoy Chakraborty +2Sparse AutoencodersMechanistic Interpretability