Interpreting and Steering LLMs with Mutual Information-based Explanations on Sparse Autoencoders

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Wu, Xuansheng, Yuan, Jiayi, Yao, Wenlin, Zhai, Xiaoming, Liu, Ninghao
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2502.15576