chore: adding HF_TOKEN

app.py

@@ -5,6 +5,8 @@ import torch
 model_name = "meta-llama/Meta-Llama-3-8B-Instruct"
 device_map = 'cuda'
 
+HF_TOKEN = os.environ.get("HF_TOKEN", None)
+
 def load_model() -> AutoModelForCausalLM:
     return AutoModelForCausalLM.from_pretrained(model_name, device_map=device_map)