Update app.py

app.py

@@ -25,7 +25,9 @@ model = LlamaForCausalLM.from_pretrained(
     model_name,
     quantization_config=quant_config
 )
-model = PeftModel.from_pretrained(model, lora_name).to("cuda")
+model = PeftModel.from_pretrained(model, lora_name)
+
+model = model.to("cuda")
 
 
 # will be used with normal template