Towards a Mechanistic Interpretation of Multi-Step Reasoning Capabilities of Language Models

Yifan Hou, Jiaoda Li, Yu Fei, Alessandro Stolfo, Wangchunshu Zhou, Guangtao Zeng, Antoine Bosselut, Mrinmaya Sachan

Submission Type: Regular Long Paper

Submission Track: Interpretability, Interactivity, and Analysis of Models for NLP

Submission Track 2: Theme Track: Large Language Models and the Future of NLP

Keywords: model interpretation, reasoning, attention mechanism, large language model

Abstract: Recent work has shown that language models (LMs) have strong multi-step (i.e., procedural) reasoning capabilities. However, it is unclear whether LMs perform these tasks by cheating with answers memorized from pretraining corpus, or, via a multi-step reasoning mechanism. In this paper, we try to answer this question by exploring a mechanistic interpretation of LMs for multi-step reasoning tasks. Concretely, we hypothesize that the LM implicitly embeds a reasoning tree resembling the correct reasoning process within it. We test this hypothesis by introducing a new probing approach (called MechanisticProbe) that recovers the reasoning tree from the model’s attention patterns. We use our probe to analyze two LMs: GPT-2 on a synthetic task (k-th smallest element), and LLaMA on two simple language-based reasoning tasks (ProofWriter \& AI2 Reasoning Challenge). We show that MechanisticProbe is able to detect the information of the reasoning tree from the model’s attentions for most examples, suggesting that the LM indeed is going through a process of multi-step reasoning within its architecture in many cases.

Submission Number: 4046