new token

model.py

@@ -5,20 +5,21 @@ import torch
 from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 
 model_id = 'meta-llama/Llama-2-13b-chat-hf'
-
+access_token='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'
 if torch.cuda.is_available():
-    config = AutoConfig.from_pretrained(model_id)
+    config = AutoConfig.from_pretrained(model_id, use_auth_token=access_token)
     config.pretraining_tp = 1
     model = AutoModelForCausalLM.from_pretrained(
         model_id,
         config=config,
         torch_dtype=torch.float16,
         load_in_4bit=True,
-        device_map='auto'
+        device_map='auto',
+        use_auth_token=access_token
     )
 else:
     model = None
-tokenizer = AutoTokenizer.from_pretrained(model_id)
+tokenizer = AutoTokenizer.from_pretrained(model_id, use_auth_token=access_token)
 
 
 def get_prompt(message: str, chat_history: list[tuple[str, str]],