A Reproduction of OpenLLaMA using 128 H100 GPUs in Bfloat16. The pretrain data consists of Falcon, Starcoder, and the wikipedia, arxiv, books, stackexchange from RedPajama. In total, this encompassed nearly 1 trillion tokens. The model was trained over a single epoch, incorporating 2000 warm-up steps and a cosine learning rate schedule, starting at 3e-5 with 4M batch size. The sole distinction from the OpenLLaMA 7B Base lies in the organization of Falcon documents, which follows the methodology outlined in this arXiv paper.
Модальности:
Генерация текста
Задача: Генерация текста
Автор: itsliupeng
Теги: llama, text-generation-inference, endpoints_compatible
Лайков: 3 | Загрузок: 73
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
