Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Ensure correct device is used for autocast when mps is selected as Fabric accelerator (#20876)
* Make sure MPS is used when chosen as accelerator in Fabric * Added mps tests to connector and Fabric --------- Co-authored-by: Haga Device <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
A
Arman Naseri committed
14b6c3e24f22aa85dceae48527ab86fdee310973
Parent: 821611b
Committed by GitHub <[email protected]>
on 6/5/2025, 5:05:38 PM