Upload app.py

app.py

@@ -81,8 +81,8 @@ def merlin_chat(message, history):
         result = model(
             prompt,
             return_full_text=False
+            max_new_tokens=250
             # truncation=True,
-            # max_length=250, 
             # handle_long_generation="hole"
             )
         response = result[0]['generated_text']