Skip to content

Fix two FP8-training issues: PLE ngram dequantization and LoRA fp8 autocast guard - #183

Open
ehooon wants to merge 2 commits into
modelscope:mainfrom
ehooon:fix-fp8-training
Open

Fix two FP8-training issues: PLE ngram dequantization and LoRA fp8 autocast guard #183
ehooon wants to merge 2 commits into
modelscope:mainfrom
ehooon:fix-fp8-training

fix(lora): exclude LoRA A/B from fp8 autocast

98bb53c
Select commit
Loading
Failed to load commit list.

Workflow runs completed with no jobs