Update app.py

app.py

@@ -22,9 +22,9 @@ model = AutoModelForCausalLM.from_pretrained(base_model_name,
                                              quantization_config=bnb_config,
                                              device_map="auto")
 model = PeftModel.from_pretrained(model, adapter_model_name, token=os.environ['HF_TOKEN'])
-model.merge_and_unload()
+model = model.merge_and_unload()
 
-model = model.to(device)
+#model = model.to(device)
 
 tokenizer = AutoTokenizer.from_pretrained(base_model_name)