Open Models Almanac
Methodsoptimizationtraining precision

single mention · optimization

BF16 attention projections

Keeping QKV and attention projections in BF16 during NVFP4-quantized pretraining.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See training precision for the methods filed beside it.

Also called keeping attention projections in BF16.

Filed under optimization › training precision