#sparse autoencoders
Sparse Autoencoders: 2 AI articles covering sparse autoencoders news, analysis, and research
Articles
How Vision Transformers Represent Position: A Mechanistic⭐9
A mechanistic study of Vision Transformers reveals how absolute and rotary positional encodings represent spatial position differently—explicit features versus ...
FailSAE: Interpretable Failure Prediction for Vision-Language⭐9
Discover how sparse autoencoders enable interpretable failure prediction in vision-language models, revealing why errors occur for risk-aware AI decisions.