more

app.py

@@ -51,6 +51,7 @@ def respond(message, max_new_tokens, temperature, system_message="You are a help
 
         # Decode the generated tokens back to text
         generated_text = tokenizer.decode(output[0], skip_special_tokens=True)
+        print("[DEBUG] Generated Text:", generated_text)
 
         # Clean up the response by removing unwanted parts (e.g., system and user info)
         cleaned_response = generated_text.split("\n")[-1]  # Assuming the response ends at the last line