SAIF: A Sparse Autoencoder Framework for Interpreting and Steering Instruction Following of Language Models

Lecture notes in computer science 2026 Open access

Read full text

Authors: Zirui He, Haiyan Zhao, Yiran Qiao, Fan Yang, Ali Payani, Jing Ma, Mengnan Du

DOI: 10.1007/978-3-032-37676-3_11

Cited by: 0

Fields of Study & Subjects

Natural Language Processing Techniques Intelligent Tutoring Systems and Adaptive Learning