Update app.py

app.py

@@ -18,7 +18,7 @@ tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
 # Load model in this way if use GPU
 if device == "cuda":
     model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True, quantization_config=bnb_config)
-else:
+elif(device == "cpu")
     model = AutoModelForCausalLM.from_pretrained("truongghieu/deci-finetuned", trust_remote_code=True)
 # Move the model to the GPU if available