423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-15 11:20:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

Bug Fix: off is parsed as False in yaml file, changed to disabled to avoid this.

Browse Source
This commit is contained in:
stceum
2024-06-24 20:39:20 +08:00
parent 5b897e7c35
commit 3ed063f281
3 changed files with 6 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

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

@@ -97,7 +97,7 @@ class ModelArguments:
default=None,
metadata={"help": "Which scaling strategy should be adopted for the RoPE embeddings."},
)
flash_attn: Literal["off", "sdpa", "fa2", "auto"] = field(
flash_attn: Literal["disabled", "sdpa", "fa2", "auto"] = field(
default="auto",
metadata={"help": "Enable FlashAttention for faster training and inference."},
)