Saved in:
Bibliographic Details
Main Authors: Lee, Andrew, Belinkov, Yonatan, Viégas, Fernanda, Wattenberg, Martin
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.04752
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Despite the central role of attention heads in Transformers, we lack tools to understand why a model attends to a particular token. To address this, we study the query-key (QK) space -- the bilinear joint embedding space between queries and keys. We present a contrastive covariance method to decompose the QK space into low-rank, human-interpretable components. It is when features in keys and queries align in these low-rank subspaces that high attention scores are produced. We first study our method both analytically and empirically in a simplified setting. We then apply our method to large language models to identify human-interpretable QK subspaces for categorical semantic features and binding features. Finally, we demonstrate how attention scores can be attributed to our identified features.