Increasing Trust in Language Models Through the Reuse of Verified Circuits

P. Quirke, C. Neo, F. Barez
February 2024
Type
Preprint
Publication
arXiv:2402.02619
Interpretability