INVALID LANGUAGE PAIR SPECIFIED. EXAMPLE: LANGPAIR=EN|IT USING 2 LETTER ISO OR RFC3066 LIKE ZH-CN. ALMOST ALL LANGUAGES SUPPORTED BUT SOME MAY HAVE NO CONTENT It has a strong focus on step-by-step problem solving and providing detailed justifications for its responses, but can also jump into roleplaying scenarios with a little nudge. This is a much more well-rounded model than my last attempt, but I’d like to keep training it on mixed datasets to continue balancing it out. It leans a little too hard toward logical problem solving sometimes. My past experiences with GPTQ + 22b models haven’t been successful so I would recommend using HF / bitsandbytes quantization for this. And an example of how it handles different roles, which I still like to use explicit instructions for:
Модальности:
Генерация текста
Задача: Генерация текста
Автор: nkpz
Теги: llama, text-generation-inference, endpoints_compatible
Лайков: 2 | Загрузок: 6
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
