Our paper Interpreting LFPs in Large Language Models has been accepted at NeurIPS 2024.