Update app.py

app.py

@@ -29,8 +29,7 @@ trained_model = AutoPeftModelForCausalLM.from_pretrained(
                 bnb_4bit_compute_dtype=torch.bfloat16,
                 bnb_4bit_use_double_quant=True,
                 bnb_4bit_quant_type='nf4'
-            ),    torch_dtype=torch.bfloat16,
-    low_cpu_mem_usage=True,
+            ),    torch_dtype=torch.bfloat16,offload_folder="offload/",
 )
 tokenizer = AutoTokenizer.from_pretrained(model_token)