Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Fix Typo in TBPTT Documentation and Improve Trainer Docstring (#20897)
K
kilavvy committed
ab619cd0e073df678ba6ba8bfbee7c3b4c67b3cb
Parent: aebf3f4
Committed by GitHub <[email protected]>
on 6/12/2025, 6:02:18 PM