Search papers, labs, and topics across Lattice.
Affiliation:
1
0
Sparse autoencoders may misrepresent human category structures, failing to capture the nuanced typicality that dense embeddings can.
