Update helper.py

helper.py

@@ -39,7 +39,7 @@ def get_fn(model_path: str, **model_kwargs):
         # Fallback to default attention implementation
         model = AutoModelForCausalLM.from_pretrained(
             model_path,
-            torch_dtype= torch.bfloat16
+            torch_dtype= torch.bfloat16,
             device_map="auto",
             quantization_config=quantization_config,
         )