diff --git a/src/axolotl/core/trainers/grpo/__init__.py b/src/axolotl/core/trainers/grpo/__init__.py
index 771f788fe..3e824f705 100644
--- a/src/axolotl/core/trainers/grpo/__init__.py
+++ b/src/axolotl/core/trainers/grpo/__init__.py
@@ -148,7 +148,7 @@ class GRPOStrategy:
 
     @classmethod
     def get_blocklist_args_kwargs(cls) -> list[str]:
-        return ["dataset_num_proc", "max_length"]
+        return ["dataset_num_proc", "max_length", "include_tokens_per_second"]
 
     @classmethod
     def get_reward_func(cls, reward_func_fqn: str) -> RewardFunc: