Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Custom DeepSpeed config format error with offload_optimizer (#17790)
Co-authored-by: francis <xiao.wei@synyi.com>
O
one-matrix committed
2af57c702a039cbb86d38cd61718fa467b346329
Parent: 3cfacc7
Committed by GitHub <noreply@github.com>
on 6/8/2023, 8:25:17 PM