Kernels:

jburtoft
/

kda-neuron-kernels

Kernel card Files Files and versions

kda-neuron-kernels / build

252 kB

Ctrl+K

Ctrl+K

5 contributors

History: 11 commits

jburtoft's picture

README: re-measured perf on torch 2.13.0; exact chunk ~1.3x; chunked-wrapper dg caveat quantified; compiled decode step; bwd_batched from ~16 items

ca4ade9 2 days ago

torch-neuron
README: re-measured perf on torch 2.13.0; exact chunk ~1.3x; chunked-wrapper dg caveat quantified; compiled decode step; bwd_batched from ~16 items 2 days ago