Update app.py

app.py

@@ -6,7 +6,7 @@ from peft import PeftModel
 # Load base model & tokenizer
 base_model = "vilsonrodrigues/falcon-7b-instruct-sharded"
 tokenizer = AutoTokenizer.from_pretrained(base_model)
-model = AutoModelForCausalLM.from_pretrained(model_name, device_map="cpu", torch_dtype=torch.float32)
+model = AutoModelForCausalLM.from_pretrained(base_model, device_map="cpu", torch_dtype=torch.float32)
 
 # Load LoRA adapter
 adapter_path = "./model"