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

fix bug

Browse Source 
Former-commit-id: b06faa1be3f5aa5e0fa31aa31314c213c36c3442
This commit is contained in:
hiyouga 2023-12-24 19:20:12 +08:00
parent 921f593632
commit c33fbea469
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()