Being authors of LLaMA is sufficient to argue they know how to train LLaMAs.