Update app.py

app.py

@@ -12,7 +12,7 @@ description = "You can use [🐋🐳microsoft/Orca-2-13b](https://huggingface.co
 # os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'max_split_size_mb:50'
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 model_name = "microsoft/Orca-2-13b"
-offload_folder = './model_weights'
+# offload_folder = './model_weights'
 
 if not os.path.exists(offload_folder):
     os.makedirs(offload_folder)
@@ -20,7 +20,7 @@ if not os.path.exists(offload_folder):
 tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False)
 model = AutoModelForCausalLM.from_pretrained(model_name)
 model = model.to(torch.bfloat16)
-model = model.to(device_map='auto', offload_folder=offload_folder)
+model = model.to(device_map='auto')
 
 class OrcaChatBot:
     def __init__(self, model, tokenizer, system_message="You are Orca, an AI language model created by Microsoft. You are a cautious assistant. You carefully follow instructions. You are helpful and harmless and you follow ethical guidelines and promote positive behavior."):