ZeRO3: Gradient norm allreduce for DP #1021
Handle Norm allreduce when no mp
a9879b89
Merge branch 'master' into olruwase/zero3_dp_norm_allreduce
3e11f4c7
jeffra
approved these changes
on 2021-04-29
tjruwase
merged
429dfa6c
into master 4 years ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub