paper-with-me

Papers

NeuronScope: A Multi-Agent Framework for Explaining Polysemantic Neurons in Language Models

2026-01-07 · Weiqi Liu, Yongliang Miao, Haiyan Zhao, Yanguang Liu, Mengnan Du arxiv

Neuron-level interpretation in large language models (LLMs) is fundamentally challenged by widespread polysemanticity, where individual neurons respond to multiple distinct semantic concepts. Existing single-pass interpretation methods struggle to faithfully capture such multi-concept behavior. In this work, we propose NeuronScope, a multi-agent framework that reformulates neuron interpretation as an iterative, activation-guided process. NeuronScope explicitly deconstructs neuron activations into atomic semantic components, clusters them into distinct semantic modes, and iteratively refines each explanation using neuron activation feedback. Experiments demonstrate that NeuronScope uncovers hidden polysemanticity and produces explanations with significantly higher activation correlation compared to single-pass baselines.

📄 PDF Abstract BibTeX arXiv:2601.03671

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

What Causes Polysemanticity? An Alternative Origin Story of Mixed Selectivity from Incidental Causes

2023-12-05 · Victor Lecomte, Kushal Thaman, Rylan Schaeffer, Naomi Bashkansky 외

Polysemantic neurons -- neurons that activate for a set of unrelated features -- have been seen as a significant obstacle towards interpretability of task-optimized deep networks, with implications for AI safety. The cla…

PURE: Turning Polysemantic Neurons Into Pure Features by Identifying Relevant Circuits

2024-04-09 · Maximilian Dreyer, Erblina Purelku, Johanna Vielhaben, Wojciech Samek 외

The field of mechanistic interpretability aims to study the role of individual neurons in Deep Neural Networks. Single neurons, however, have the capability to act polysemantically and encode for multiple (unrelated) fea…

Probing the Vulnerability of Large Language Models to Polysemantic Interventions

2025-05-16 · Bofan Gong, Shiyang Lai, Dawn Song

Polysemanticity -- where individual neurons encode multiple unrelated features -- is a well-known characteristic of large neural networks and remains a central challenge in the interpretability of language models. At the…

Disentangling Polysemantic Channels in Convolutional Neural Networks

2025-04-17 · Robin Hesse, Jonas Fischer, Simone Schaub-Meyer, Stefan Roth

Mechanistic interpretability is concerned with analyzing individual components in a (convolutional) neural network (CNN) and how they form larger circuits representing decision mechanisms. These investigations are challe…

Understanding polysemanticity in neural networks through coding theory

2024-01-31 · Simon C. Marshall, Jan H. Kirchner

Despite substantial efforts, neural network interpretability remains an elusive goal, with previous research failing to provide succinct explanations of most single neurons' impact on the network output. This limitation …