Model techniques map
Techniquesoptimizationtraining precision

implementation detail · filed under optimization

FP32 gradient reduction

Gradient reduction uses FP32 precision, restoring the full-precision reduction recipe.

source
1
model
1
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

Rolling back to an earlier checkpoint and moving back to the full FP32 gradient reduction recipe re-stabilized training

usedoptimizationin Nemotron 3 UltraNVIDIA

Filed alongside

Other methods under optimization :: training precision.