Show HN: HyperSAE – Hyperbolic Sparse Autoencoders for LLM Interpretabilitygithub.com3 points·visha1v··0 commentsOpen articleSaveView on HN