Update app.py

app.py

@@ -21,8 +21,9 @@ def respond(message, history):
 
     # encode + generate
     inputs  = tokenizer(prompt, return_tensors="pt")
+    print(**inputs)
     outputs = model.generate(
-        inputs[0],
+        **inputs,
         max_new_tokens=200,
         do_sample=False,     # greedy
         pad_token_id=tokenizer.eos_token_id