Singular Vectors of Attention Heads Align with Features.
Technical Report. doi:10.48550/arXiv.2602.13524
Mechanistic Interpretability |
| [Franco et al., 2026] | Gabriel Franco, Carson Loughridge and Mark Crovella (2026). Singular Vectors of Attention Heads Align with Features. Technical Report. doi:10.48550/arXiv.2602.13524 |
| [Franco et al., 2026] | Gabriel Franco, Lucas M. Tassis, Azalea Rohr and Mark Crovella (2026). Finding Highly Interpretable Prompt-Specific Circuits in Language Models. Technical Report. doi:10.48550/arXiv.2602.13483 |
| [Franco and Crovella, 2025] | Gabriel Franco and Mark Crovella (2025). Pinpointing Attention-Causal Communication in Language Models. In: Proceedings of NeurIPS. San Diego, CA. Also appeared in Mechanistic Interpretability Workshop at NeurIPS 2025. doi:TBD |
| [Franco and Crovella, 2024] | Gabriel Franco and Mark Crovella (2024). Sparse Attention Decomposition Applied to Circuit Tracing. Technical Report Nr. 2410.00340. doi:10.48550/arXiv.2410.00340 |