Models for (steerable) story-writing and role-playing. Other quants Other models — Opus V1 prompting guide with many prompts and interactive examples you can copy. — Story-writing examples (model output in red) — Role-playing examples (model output in red, including the user turns) — Join the community on Discord to discuss the models. This model uses an extended version of the Llama 3 template (click for details). Compared to the official Llama 3 template, a new writer role was added. — The writer messages can (but don’t have to have) character:, these are used to indicate the «active» character during role-play. — There can be multiple subsequent message with a writer role, especially if names are involved. While the main goal for the models is great story-writing and role-playing performance, the models are also capable of several writing related tasks as well as general assistance. Read the official guide to learn how to prompt the model for various tasks beyond story-writing and role-playing, such as summarization, plot idea generation and more. For story-writing and role-play, I recommend «Min P» based sampling with minp in the range [0.01, 0.1] and with temperature in the…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: dreamgen
Теги: llama, unsloth, axolotl, conversational, en, text-generation-inference, endpoints_compatible, 6-bit
Лайков: 3 | Загрузок: 18
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
