Update helper.py

helper.py

@@ -114,7 +114,7 @@ def registry(name: str = None, **kwargs) -> gr.ChatInterface:
     predict_fn = get_fn(model_name=name, **kwargs)
     
     # Define a wrapper function that integrates preprocessing and postprocessing
-    def wrapper(message, history, system_prompt, temperature, max_tokens, top_k, repetition_penalty, top_p):
+    def wrapper(message, history, system_prompt, temperature, max_tokens, top_p):
         # Preprocess the inputs
         preprocessed = preprocess(message, history)
         
@@ -132,7 +132,7 @@ def registry(name: str = None, **kwargs) -> gr.ChatInterface:
         # Collect the generated response
         response = ""
         for partial_response in response_generator:
-            response = partial_response  # Gradio will handle streaming
+            response += partial_response  # Gradio will handle streaming
             yield response
 
     # Create the Gradio ChatInterface with the wrapper function