arXiv 2610.21261 · cs.LG · 2026-10-15
Behavioral Compositionality of Sparse Autoencoder Representations
Haotian Ma, Ziye Mei, Frederic Sala
OpenSAE
In latest · #133 Added on 2026-10-21
One-line takeaway
Introduces SAR to measure whether sparse-autoencoder features compose additively in downstream behaviour, aiding interpretability.
提出 SAR 指标，衡量稀疏自编码器特征在下游行为上能否“相加组合”，服务可解释性研究。
Read on arXiv PDF Hugging Face discussion Code Project page
Topics & disciplines
More on Safety & interpretability
- DecepEval: A Benchmark for Evaluating Deception in LLM Agents ▲85
- Lexical Reference Frames: Interpreting LM Residual Streams in Native Semantics ▲55
- U-Space: Uncovering When and Why Uncertainty Arises in Language Models ▲54
- Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures ▲29
- APEX-Harm: Evaluating Agent Safety in Shared Professional Workspaces ▲26
About this data
Metadata from the arXiv API (verified); upvotes from Hugging Face Daily Papers; citations from Semantic Scholar. Last refreshed 2026-10-22.
The takeaway is written by AIMap from the abstract; the abstract itself is not reproduced — read the original.