Context Copying Modulation: The Role of Entropy Neurons in Managing Parametric and Contextual Knowledge Conflicts

ACL ARR 2025 May Submission3536 Authors

Abstract: The behavior of Large Language Models (LLMs) when facing contextual information that conflicts with their internal parametric knowledge is inconsistent, with no generally accepted explanation for the expected outcome distribution. Recent work has identified in autoregressive transformer models a class of neurons -- called $\textit{entropy neurons}$ -- that produce a significant effect on the model output entropy while having an overall moderate impact on the ranking of the predicted tokens. In this paper, we investigate the preliminary claim that these neurons are involved in inhibiting context copying behavior in transformers by looking at their role in resolving conflicts between contextual and parametric information. We show that $\textit{entropy neurons}$ are responsible for suppressing context copying across a range of LLMs, and that ablating them leads to a significant change in the generation process. These results enhance our understanding of the internal dynamics of LLMs when handling conflicting information.

Paper Type: Short

Research Area: Interpretability and Analysis of Models for NLP

Research Area Keywords: adversarial attacks/examples/training, counterfactual/contrastive explanations, knowledge tracing/discovering/inducing, probing

Contribution Types: Model analysis & interpretability

Languages Studied: English

Submission Number: 3536