Update models.py

models.py

@@ -133,12 +133,14 @@ login(token=hf_api_key)
 
 
 
-
 model_id = "meta-llama/Meta-Llama-3.1-8B-Instruct"
+model_config = transformers.LlamaConfig.from_pretrained(model_id)
+model_config.rope_scaling = {"type": "linear", "factor": 1.0}
+model = transformers.AutoModelForCausalLM.from_pretrained(model_id, config=model_config)
 
 pipeline = transformers.pipeline(
     "text-generation",
-    model=model_id,
+    model=model,
     model_kwargs={"torch_dtype": torch.bfloat16},
     device_map="auto",
 )