Towards Interpreting Language Models: A Case Study in Multi-Hop Reasoning

Understand

Answering multi-hop reasoning questions requires retrieving and synthesizing information from diverse sources.

• Language models (LMs) struggle to perform such reasoning consistently.
• We propose an approach to pinpoint and rectify multi-hop reasoning failures through targeted memory injections on LM attention heads.
• First, we analyze the per-layer activations of GPT-2 models in response to single- and multi-hop prompts.