Update app.py

app.py

@@ -4,8 +4,10 @@ from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
 
 model_name = "anasmkh/customized_llama3.1_8b"
 
+from transformers import AutoModel
+model = AutoModel.from_pretrained("anasmkh/customized_llama3.1_8b",torch_dtype=torch.float16)
 tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype=torch.float16)
+# model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype=torch.float16)
 
 generator = pipeline(
     "text-generation",