423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2026-08-17 21:45:48 +08:00
Code Issues Packages Projects Releases Wiki Activity

[feat] support HyperParallel PT training and activation optimization (#10370)

Browse Source
This commit is contained in:
Cui-yshoho
2026-06-02 22:39:32 +08:00
committed by GitHub
parent a98a1ef101
commit 053d43c0ac
5 changed files with 326 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/llamafactory/hparams/finetuning_args.py
View File

@@ -487,7 +487,7 @@ class FinetuningArguments(
metadata={
"help": (
"Whether or not to use HyperParallel distributed training backend (FSDP/TP). "
"Only supported for the 'sft' stage with full fine-tuning."
"Only supported for the 'pt' and 'sft' stages with full fine-tuning."
)
},
)