基于SAE的机制可解释性：探究大语言模型中的宗教、暴力与地理
Mechanistic Interpretability with SAEs: Probing Religion, Violence, and Geography in Large Language Models
AI总结：
本文利用稀疏自编码器对五种大语言模型进行机制可解释性分析，发现各宗教内部表征一致性相近，但伊斯兰教更易关联暴力语言特征，而地理关联主要反映真实宗教人口分布，揭示了模型同时嵌入事实与文化刻板印象。
摘要：
尽管关于大语言模型（LLMs）偏见的研究日益增多，但大多数工作聚焦于性别和种族，对宗教身份的关注很少。本文探讨了宗教如何在LLMs内部被表示，以及它如何与暴力和地理概念相交织。利用机制可解释性和通过Neuronpedia API使用的稀疏自编码器（SAEs），我们分析了五个模型中潜在特征激活情况。我们测量了宗教相关提示与暴力相关提示之间的重叠，并探查了激活上下文中的语义模式。虽然所有五种宗教都表现出相当的内部一致性，但伊斯兰教更频繁地与暴力语言相关的特征联系在一起。相比之下，地理关联在很大程度上反映了现实世界的宗教人口分布，揭示了模型如何同时嵌入事实分布和文化刻板印象。这些发现凸显了结构分析在审计中的价值，即不仅审计输出，还审计塑造模型行为的内部表示。
英文摘要：
Despite growing research on bias in large language models (LLMs), most work has focused on gender and race, with little attention to religious identity. This paper explores how religion is internally represented in LLMs and how it intersects with concepts of violence and geography. Using mechanistic interpretability and Sparse Autoencoders (SAEs) via the Neuronpedia API, we analyze latent feature activations across five models. We measure overlap between religion- and violence-related prompts and probe semantic patterns in activation contexts. While all five religions show comparable internal cohesion, Islam is more frequently linked to features associated with violent language. In contrast, geographic associations largely reflect real-world religious demographics, revealing how models embed both factual distributions and cultural stereotypes. These findings highlight the value of structural analysis in auditing not just outputs but also internal representations that shape model behavior.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
