"attention head analysis" Papers
3 papers found
Conference
Efficient Automated Circuit Discovery in Transformers using Contextual Decomposition
Aliyah Hsu, Georgia Zhou, Yeshwanth Cherapanamjeri et al.
ICLR 2025arXiv:2407.00886
15
citations
Unveiling Visual Perception in Language Models: An Attention Head Analysis Approach
Jing Bi, Lianggong Bruce Wen, Zhang Liu et al.
CVPR 2025arXiv:2412.18108
18
citations
Interpreting and Improving Large Language Models in Arithmetic Calculation
Wei Zhang, Wan Chaoqun, Yonggang Zhang et al.
ICML 2024arXiv:2409.01659
42
citations