Open Models Almanac
Methodsmodel architecturecontext capacity

single mention · model architecture

Multi-head hashing for N-gram embedding lookup

Multiple hash heads are used to index N-gram embeddings.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See context capacity for the methods filed beside it.

Also called multi-head hashing.

Filed under model architecture › context capacity