Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
fix: always call `WandbLogger.experiment` first in `_call_setup_hook` to ensure `tensorboard` logs can sync to `wandb` (#20610)
(cherry picked from commit 93d707df4c130046cf4a5905f469f62f143ecf38)
M
Matthew Hoffman committed
4423f18f53b95571195c62d8f058c2d35ee47906
Parent: a11f601
Committed by lexierule <75488056+lexierule@users.noreply.github.com>
on 3/18/2025, 9:08:54 PM