Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Formal Algorithms for Transformers. Parent comment
This is ironic considering they use more words to describe chunking (splitting along a dimension, x,y = a[0,:,:,:], a[1,:,:,:]) than multi-heads.
There are no comments on this comment yet.
This is ironic considering they use more words to describe chunking (splitting along a dimension, x,y = a[0,:,:,:], a[1,:,:,:]) than multi-heads.