zhangyiss/mlx-examples
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx-examples.git synced 2025-12-16 02:08:55 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

updates

Browse Source
This commit is contained in:
Goekdeniz-Guelmez
2025-03-05 14:49:56 +01:00
parent 9a36452519
commit d723ddfeda
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
llms/mlx_lm/tuner/grpo_trainer.py
View File

@@ -343,6 +343,7 @@ def grpo_loss(
# Convert to tensor # Convert to tensor
prompt_tensor = mx.array(padded_prompts) prompt_tensor = mx.array(padded_prompts)
prompt_tensor = mx.stop_gradient(prompt_tensor) # Explicitly stop gradient on input
try: try:
mx.metal.clear_cache() mx.metal.clear_cache()