- 02 Apr, 2022 1 commit
-
-
ver217 authored
-
- 01 Apr, 2022 24 commits
-
-
HELSON authored
-
KAIYUAN GAN authored
-
アマデウス authored
-
アマデウス authored
-
アマデウス authored
-
アマデウス authored
-
アマデウス authored
-
アマデウス authored
-
アマデウス authored
-
アマデウス authored
-
アマデウス authored
-
Ziyue Jiang authored
-
ver217 authored
-
ver217 authored
-
ver217 authored
-
ver217 authored
-
ver217 authored
-
FredHuang99 authored
-
ver217 authored
-
LuGY authored
-
ver217 authored
* fix sharded optim zero grad * polish comments
-
アマデウス authored
-
アマデウス authored
-
Jiarui Fang authored
-
- 31 Mar, 2022 9 commits
-
-
ver217 authored
-
HELSON authored
* support existing sharded and unsharded parameters in zero * add unitest for moe-zero model init * polish moe gradient handler
-
LuGY authored
-
Wesley authored
-
Wesley authored
-
BoxiangW authored
Change the clang-format style to google style
-
ver217 authored
-
Jiarui Fang authored
-
Liang Bowen authored
-
- 30 Mar, 2022 6 commits
-
-
Jiarui Fang authored
-
LuGY authored
-
ver217 authored
* hijack p.grad in sharded model * polish comments * polish comments
-
Jiarui Fang authored
-
github-actions[bot] authored
-
Jiarui Fang authored
-