Sparse Autoencoders Learn Monosemantic Features in Vision-Language ModelsPublished byqbouniot.github.ioon •1 min readMateusz Pach, Shyamgopal Karthik, Quentin Bouniot, Serge Belongie, Zeynep AkataSparseAutoencodersLearnMonosemanticFeaturesLearn moreShareLegalReport
