Update app.py

app.py

@@ -16,7 +16,7 @@ def predict(user_input):
     inputs = tokenizer(user_input, return_tensors="tf", padding=True, truncation=True)
 
     # Generate the response using the model
-    response_ids = model.generate(
+    response_id = model.generate(
         inputs['input_ids'],
         max_length=128,         # Set max length of response
         do_sample=True,         # Sampling for variability