Saved in:
Bibliographic Details
Main Authors: He, Zhengfu, Wang, Junxuan, Lin, Rui, Ge, Xuyang, Shu, Wentao, Tang, Qiong, Zhang, Junping, Qiu, Xipeng
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.20938
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909597103554560
author He, Zhengfu
Wang, Junxuan
Lin, Rui
Ge, Xuyang
Shu, Wentao
Tang, Qiong
Zhang, Junping
Qiu, Xipeng
author_facet He, Zhengfu
Wang, Junxuan
Lin, Rui
Ge, Xuyang
Shu, Wentao
Tang, Qiong
Zhang, Junping
Qiu, Xipeng
contents We propose Low-Rank Sparse Attention (Lorsa), a sparse replacement model of Transformer attention layers to disentangle original Multi Head Self Attention (MHSA) into individually comprehensible components. Lorsa is designed to address the challenge of attention superposition to understand attention-mediated interaction between features in different token positions. We show that Lorsa heads find cleaner and finer-grained versions of previously discovered MHSA behaviors like induction heads, successor heads and attention sink behavior (i.e., heavily attending to the first token). Lorsa and Sparse Autoencoder (SAE) are both sparse dictionary learning methods applied to different Transformer components, and lead to consistent findings in many ways. For instance, we discover a comprehensive family of arithmetic-specific Lorsa heads, each corresponding to an atomic operation in Llama-3.1-8B. Automated interpretability analysis indicates that Lorsa achieves parity with SAE in interpretability while Lorsa exhibits superior circuit discovery properties, especially for features computed collectively by multiple MHSA heads. We also conduct extensive experiments on architectural design ablation, Lorsa scaling law and error analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2504_20938
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Understanding the Nature of Attention with Low-Rank Sparse Decomposition
He, Zhengfu
Wang, Junxuan
Lin, Rui
Ge, Xuyang
Shu, Wentao
Tang, Qiong
Zhang, Junping
Qiu, Xipeng
Machine Learning
Computation and Language
We propose Low-Rank Sparse Attention (Lorsa), a sparse replacement model of Transformer attention layers to disentangle original Multi Head Self Attention (MHSA) into individually comprehensible components. Lorsa is designed to address the challenge of attention superposition to understand attention-mediated interaction between features in different token positions. We show that Lorsa heads find cleaner and finer-grained versions of previously discovered MHSA behaviors like induction heads, successor heads and attention sink behavior (i.e., heavily attending to the first token). Lorsa and Sparse Autoencoder (SAE) are both sparse dictionary learning methods applied to different Transformer components, and lead to consistent findings in many ways. For instance, we discover a comprehensive family of arithmetic-specific Lorsa heads, each corresponding to an atomic operation in Llama-3.1-8B. Automated interpretability analysis indicates that Lorsa achieves parity with SAE in interpretability while Lorsa exhibits superior circuit discovery properties, especially for features computed collectively by multiple MHSA heads. We also conduct extensive experiments on architectural design ablation, Lorsa scaling law and error analysis.
title Towards Understanding the Nature of Attention with Low-Rank Sparse Decomposition
topic Machine Learning
Computation and Language
url https://arxiv.org/abs/2504.20938