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


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.
Anthology ID:
2023.emnlp-main.299
Volume:
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4902–4919
Language:
URL:
https://aclanthology.org/2023.emnlp-main.299
DOI:
10.18653/v1/2023.emnlp-main.299
Bibkey:
Cite (ACL):
Yifan Hou, Jiaoda Li, Yu Fei, Alessandro Stolfo, Wangchunshu Zhou, Guangtao Zeng, Antoine Bosselut, and Mrinmaya Sachan. 2023. Towards a Mechanistic Interpretation of Multi-Step Reasoning Capabilities of Language Models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 4902–4919, Singapore. Association for Computational Linguistics.
Cite (Informal):
Towards a Mechanistic Interpretation of Multi-Step Reasoning Capabilities of Language Models (Hou et al., EMNLP 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-2024-clasp/2023.emnlp-main.299.pdf
Video:
 https://preview.aclanthology.org/ingest-2024-clasp/2023.emnlp-main.299.mp4