Sparse autoencoders uncover biologically interpretable features in protein language model representations

Journal:
Proceedings of the National Academy of Sciences of the United States of America
Published:
DOI:
10.1073/pnas.2506316122
Affiliations:
4
Authors:
4
Go to article
Institutions Authors Share
Massachusetts Institute of Technology (MIT), United States of America (USA)
4.000000
 1.00
Interested in Biochemistry and Cell Biology?