Separating Interpretability and Reconstruction in Sparse Autoencoders
arXiv:2611.25165v1 Announce Type: new Abstract: Sparse Autoencoders (SAEs) are effective tools for mechanistic interpretability by decomposing neural representations into sparse and interpretable features. However, a single sparse representation in traditional SAEs is responsible for both reconstruction fidelity and interpretability, which are not guaranteed to align. To separate these objectives, we propose DualSAE, a modular framework that augments standard SAEs with a plug-in Residual Branch. A standard SAE serves as the Sparse Branch, learning a sparse code that contains the dominant seman
阅读原文 · arXiv 机器学习本文由 SoccerDream AI 自动聚合索引,内容版权归原作者与 arXiv 机器学习 所有。
本站仅作技术动态聚合,不构成任何形式的建议。
本站仅作技术动态聚合,不构成任何形式的建议。