Update app.py

app.py

@@ -115,7 +115,7 @@ def load_model(model_path, progress=gr.Progress()):
 def generate_response_non_streaming(instruction, model_name, temperature=0.7, max_tokens=1024):
     """Generate a response from the Apollo model (non-streaming)"""
     global current_model, current_tokenizer, current_model_path
-    
+    print("instruction:",instruction)
     # If model is not yet loaded, load it first
     if current_model_path != model_name or current_model is None:
         load_message = load_model(model_name)