423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-10-16 00:28:10 +08:00
Code Issues Packages Projects Releases Wiki Activity

tiny fix

Browse Source 
Former-commit-id: 19e43c3a9ed771e991cb273d394ab28fb923f868
This commit is contained in:
hiyouga 2024-07-01 03:55:20 +08:00
parent 884b49e662
commit 4357e42391
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/llamafactory/model/model_utils/attention.py
View File

@ -35,7 +35,7 @@ def configure_attn_implementation(
if model_args.flash_attn == "auto": if model_args.flash_attn == "auto":
logger.warning("Gemma-2 models should use eager attention in training, change `flash_attn` to disabled.") logger.warning("Gemma-2 models should use eager attention in training, change `flash_attn` to disabled.")
model_args.flash_attn = "disabled" model_args.flash_attn = "disabled"
else: elif model_args.flash_attn != "disabled":
logger.warning( logger.warning(
"Gemma-2 models should use eager attention in training, but you set `flash_attn: {}`. " "Gemma-2 models should use eager attention in training, but you set `flash_attn: {}`. "
"Will proceed at your own risk.".format(model_args.flash_attn) "Will proceed at your own risk.".format(model_args.flash_attn)