Towards Free Normalization: Fusing Normalization into GEMM and Attention Kernels (pytorch.org)

1 points by matt_d 3d ago ↗ HN

0 comments

[ 2.9 ms ] story [ 14.4 ms ] thread

No comments yet.

A showcase of shard-db. Data from the Hacker News API. Refreshed every 15 minutes.

Source · Live DB stats