This model is a fine-tuned version of microsoft/Orca-2-13b on a subset of the Vezora/MiniOrcaUncencored_Alpaca dataset, with some particularly spicy prompts added as well, to reduce the risk of rejections. The following hyperparameters were used during training: — learningrate: 0.0002 — trainbatchsize: 6 — evalbatchsize: 6 — seed: 42 — gradientaccumulationsteps: 6 — totaltrainbatchsize: 36 — optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 — lrschedulertype: cosine — lrschedulerwarmupsteps: 15 — numepochs: 2
Модальности:
Генерация текста
Задача: Генерация текста
Автор: athirdpath
Теги: llama, generated_from_trainer, endpoints_compatible, 4-bit, bitsandbytes
Лайков: 2 | Загрузок: 13
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.