|
@@ -90,7 +90,7 @@ class ModelTrainer:
|
|
)
|
|
)
|
|
|
|
|
|
torch.cuda.empty_cache()
|
|
torch.cuda.empty_cache()
|
|
-
|
|
|
|
|
|
+ print("self.config.learning_rate=",self.config.learning_rate)
|
|
training_args = GRPOConfig(
|
|
training_args = GRPOConfig(
|
|
use_vllm=self.config.use_vllm,
|
|
use_vllm=self.config.use_vllm,
|
|
learning_rate=self.config.learning_rate,
|
|
learning_rate=self.config.learning_rate,
|