Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
docs: Add float16 precision warning for optimizer eps and small values (#21295)
* docs: add warnings about Float16 limitations and introduce BFloat16 for better stability * Empty-Commit --------- Co-authored-by: jirka <[email protected]>
B
Bhimraj Yadav committed
56204d8a9dd480503c09c1d731bd16a45a8335e5
Parent: 29ddf82
Committed by GitHub <[email protected]>
on 10/21/2025, 12:45:50 PM