Skip to content

Support distributed Adam with T5 and support overlapped grad reductions with pipeline parallelism#4900

Merged
ericharper merged 30 commits intoNVIDIA-NeMo:mainfrom
timmoon10:dist-adam-pipeline-parallel-async-grad-reduction
Oct 21, 2022
Merged

Support distributed Adam with T5 and support overlapped grad reductions with pipeline parallelism#4900
ericharper merged 30 commits intoNVIDIA-NeMo:mainfrom
timmoon10:dist-adam-pipeline-parallel-async-grad-reduction

Commits

Commits on Aug 23, 2022

Commits on Aug 25, 2022

Commits on Aug 26, 2022

Commits on Aug 29, 2022

Commits on Sep 7, 2022

Commits on Sep 8, 2022

Commits on Sep 9, 2022

Commits on Sep 15, 2022

Commits on Sep 20, 2022

Commits on Oct 19, 2022

Commits on Oct 20, 2022

Comments