Making large AI models cheaper, faster and more accessible
[zero] test gradient accumulation (#1964)
* [zero] fix memory leak for zero2 * [zero] test gradient accumulation * [zero] remove grad clip test
H
HELSON committed
a1ce02d740e5013a14225620d60133104a07e8fa
Parent: b0936e4
Committed by GitHub <noreply@github.com>
on 11/29/2022, 5:00:30 AM