Beyond Top Activations: Efficient and Reliable Crowdsourced Evaluation of Automated Interpretability

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Oikarinen, Tuomas, Yan, Ge, Kulkarni, Akshay, Weng, Tsui-Wei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!