Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models
Marks, Samuel, Can Rager, Eric J. Michaud, Yonatan Belinkov, David Bau, Aaron Mueller
De quem é
Northeastern Universityafiliação declarada pelos autores, segundo o OpenAlex
Outros papers das mesmas organizações
- Kolmogorov-Arnold Networks Meet ScienceHarvard University · 687 citações
- Mental-LLM: Leveraging Large Language Models for Mental Health Prediction via Online Text DataNortheastern University · 658 citações
- Unified Concept Editing in Diffusion ModelsNortheastern University · 546 citações
- VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality DocumentsNortheastern University · 297 citações
- Improving LoRA in Privacy-preserving Federated LearningNortheastern University · 245 citações