Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals

arXiv:2402.11655v1 Announce Type: new
Abstract: Interpretability research aims to bridge the gap between the empirical success and our scientific understanding of the inner workings of large language models (LLMs). However, most existing research in this area focused on analyzing a single mechanism, such as how models copy or recall factual knowledge. In this work, we propose the formulation of competition of mechanisms, which instead of individual mechanisms focuses on the interplay of multiple mechanisms, and traces how one of them …

arxiv competition cs.cl facts language language models tracing type