AI RESEARCH
Multi-component Causal Tracing in Large Language Models
arXiv CS.LG
•
ArXi:2606.03085v1 Announce Type: new Causal tracing systematically intervenes on a large language model's (LLM's) internal representations to uncover and quantify the causal pathways linking specific inputs or computations to specific metrics of interest, quantifying the LLM's behavior. Building on previous single-component or single-layer studies, this paper presents a unified framework for causally tracing multiple components simultaneously.