Continuation of https://huggingface.co/nkpz/llama2-22b-daydreamer-v1 Trained it a couple more epochs on airoboros v2 general data. This resulted in improvements to its obsession with overexplaining itself — it’s a little less instruction focused and feels more versatile. Still needs work overall, but I think it’s moving in the right direction. Still more to come. Going to try to find a way to improve its problem solving skills without ruining the output style.
Модальности:
Генерация текста
Задача: Генерация текста
Автор: nkpz
Теги: llama, text-generation-inference, endpoints_compatible
Лайков: 2 | Загрузок: 8
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
