Update app.py

app.py

@@ -18,6 +18,7 @@ def llama2_chat(prompt):
         input_ids=inputs["input_ids"],
         attention_mask=inputs["attention_mask"],  # Pass attention_mask!
         max_new_tokens=100,
+        max_length=30,
         temperature=0.3
     )
     response = tokenizer.decode(output[0], skip_special_tokens=True)