423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-17 20:30:36 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix #5048

Browse Source 
Former-commit-id: 71a6861667ae68c1fd6a69acf68e1359b858cf1b
This commit is contained in:
hiyouga
2024-08-05 23:48:19 +08:00
parent 2e477e7458
commit 13093963b1
13 changed files with 111 additions and 69 deletions
Download Patch File Download Diff File Expand all files Collapse all files

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

@@ -36,7 +36,7 @@ def configure_attn_implementation(
if model_args.flash_attn == "auto" or model_args.flash_attn == "fa2":
if is_flash_attn_2_available():
require_version("transformers>=4.42.4", "To fix: pip install transformers>=4.42.4")
require_version("flash_attn>=2.6.0", "To fix: pip install flash_attn>=2.6.0")
require_version("flash_attn>=2.6.3", "To fix: pip install flash_attn>=2.6.3")
logger.warning("Gemma-2 should use flash attention 2, change `flash_attn` to fa2.")
model_args.flash_attn = "fa2"
else: