On the transferability of Sparse Autoencoders for interpreting compressed models
"解释性" 相关文章
PLEX: Perturbation-free Local Explanations for LLM-Based Text Classification
Foundation versus Domain-specific Models: Performance Comparison, Fusion, and Explainability in Face Recognition
Source Attribution in Retrieval-Augmented Generation
Against blanket arguments against interpretability
Activation space interpretability may be doomed
(Maybe) A Bag of Heuristics is All There Is & A Bag of Heuristics is All You Need
Deceptive agents can collude to hide dangerous features in SAEs
Mapping Neural Networks to Graph Structures: Enhancing Model Selection and Interpretability through Network Science
