Update app.py

app.py

@@ -9,7 +9,7 @@ from peft import PeftModel
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
 """
 #client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-from util.llama_rope_scaled_monkey_patch import replace_llama_rope_with_scaled_rope
+from llama_rope_scaled_monkey_patch import replace_llama_rope_with_scaled_rope
         replace_llama_rope_with_scaled_rope()
 base_model = "Neko-Institute-of-Science/LLaMA-65B-HF"
 lora_weights = "adapter_config.json"