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 bug

Browse Source 
Former-commit-id: e4bb846c43c087bfdd99d0c9eb0318e95b943977
This commit is contained in:
hiyouga 2023-12-24 19:20:12 +08:00
parent 2a3980d6ba
commit 498b83d77a
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

@ -199,7 +199,7 @@ def _prepare_model_for_training(
logger.info("Upcasting layernorm weights in float32.") logger.info("Upcasting layernorm weights in float32.")
if not model_args.disable_gradient_checkpointing: if not model_args.disable_gradient_checkpointing:
if getattr(model, "supports_gradient_checkpointing", False): if not getattr(model, "supports_gradient_checkpointing", False):
logger.warning("Current model does not support gradient checkpointing.") logger.warning("Current model does not support gradient checkpointing.")
else: else:
model.enable_input_require_grads() model.enable_input_require_grads()