← 研踪 PaperTrailACL Anthology · ACL 2026:acl:2026.acl-srw.84 · 记录 #1
Mechanistic Analysis Of Universality: Numerical Comparison Circuits Across Transformer Architectures
来源日期（仅到月份） 2026-07
来源记录日期 2026-07
雷达观测 2026-11-01 21:44
摘要
Transformer language models reliably achieve high accuracy on many reasoning tasks; however, their internal mechanisms are not fully understood. Mechanistic interpretability seeks to remedy this gap by identifying task circuits within individual models, but it is unclear whether such circuits generalize across model families and scales. In this work, we study the universality of circuits through the lens of numerical comparisons, a simple and controlled task that enables clean and causal interventions. We conduct experiments on a set of transformer models spanning different families and sizes from 1.7b to 9b parameters. We find that models within the Qwen family exhibit a highly consistent circuit structure across architecture and scale, featuring localized attention heads that write a task relevant signal. In contrast, models from other families show qualitatively different implementations, where task relevant information emerges much earlier and is distributed across components as opposed to being concentrated within a small set of attention heads. These results serve as evidence that task behavior similarities do not imply mechanistic universality and highlight the necessity for cross model comparisons to claim generalization of internal circuits.
论文发现出处
这些页面提供论文线索，不代表同行评审或质量认证。
观测到的版本
平台记录号代表本站观测到的元数据变化，不是作者修订号。来源日期仅到月份或年份时，按该月或该年首日存档。
记录 #1 · 2026-07 · 最新观测版本
Mechanistic Analysis Of Universality: Numerical Comparison Circuits Across Transformer Architectures
Transformer language models reliably achieve high accuracy on many reasoning tasks; however, their internal mechanisms are not fully understood. Mechanistic interpretability seeks to remedy this gap by identifying task circuits within individual models, but it is unclear whether such circuits generalize across model families and scales. In this work, we study the universality of circuits through the lens of numerical comparisons, a simple and controlled task that enables clean and causal interventions. We conduct experiments on a set of transformer models spanning different families and sizes from 1.7b to 9b parameters. We find that models within the Qwen family exhibit a highly consistent circuit structure across architecture and scale, featuring localized attention heads that write a task relevant signal. In contrast, models from other families show qualitatively different implementations, where task relevant information emerges much earlier and is distributed across components as opposed to being concentrated within a small set of attention heads. These results serve as evidence that task behavior similarities do not imply mechanistic universality and highlight the necessity for cross model comparisons to claim generalization of internal circuits.
查看此版本解读 →查看此版本原文 ↗