in llama.cpp load gguf has some issue in it

[wac81]

check this and how to fix it in llama.cpp: https://github.com/ggerganov/llama.cpp/issues/4772