"Uncovering Safety Risks in Open-source LLMs through Concept Activation Vector."

Zhihao Xu et al. (2024)

Details and statistics

DOI: 10.48550/ARXIV.2404.12038

access: open

type: Informal or Other Publication

metadata version: 2024-05-22

a service of  Schloss Dagstuhl - Leibniz Center for Informatics