423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-08-04 12:42:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

🐞 fix: typo

Browse Source 
Former-commit-id: b06a31e76ac623ea89ecc567dd9e9ec88dfdb585
This commit is contained in:
A-Cepheus 2024-01-22 16:04:39 +08:00
parent 39d9aba166
commit f00ad6b4f8
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/llmtuner/model/patcher.py
View File

@ -284,7 +284,7 @@ def patch_model(
if is_trainable: if is_trainable:
_prepare_model_for_training(model, model_args) _prepare_model_for_training(model, model_args)
if getattr(config, "model_type", None) == "mixtral" and is_deepspeed_zero3_enabled(): if getattr(model.config, "model_type", None) == "mixtral" and is_deepspeed_zero3_enabled():
from deepspeed.utils import set_z3_leaf_modules from deepspeed.utils import set_z3_leaf_modules
from transformers.models.mixtral.modeling_mixtral import MixtralSparseMoeBlock from transformers.models.mixtral.modeling_mixtral import MixtralSparseMoeBlock
set_z3_leaf_modules(model, [MixtralSparseMoeBlock]) set_z3_leaf_modules(model, [MixtralSparseMoeBlock])