arXiv 2610.21261 · cs.LG · 2026-10-15
Behavioral Compositionality of Sparse Autoencoder Representations
Haotian Ma, Ziye Mei, Frederic Sala
OpenSAE
在最新榜 #133 本站收录于 2026-10-21
一句话要点
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。
Introduces SAR to measure whether sparse-autoencoder features compose additively in downstream behaviour, aiding interpretability.
阅读 arXiv 原文 PDF Hugging Face 讨论 代码仓库 项目页
主题与学科
同主题：安全与可解释
- 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
数据说明
元数据来自 arXiv API（已核实）；点赞来自 Hugging Face 每日论文；引用来自 Semantic Scholar。最近刷新 2026-10-22。
要点由本站根据摘要自写，不转载摘要原文，请阅读原文。