Inoichan/GIT-Llama-2-7B

Welcome to the GIT-LLM repository. GIT-LLM is an innovative fusion of the GIT Vision and Language model with the linguistic capabilities of the LLM (Language Learning Model). Harnessing the power of both worlds, this model is fine-tuned using the LoRA (Local Re-Attention) method, optimizing it for enhanced performance in diverse vision and language tasks. This model was trained for one epoch using M3IT (excluding videos and Chinese tasks). For more details, please refer to our blog (in Japanese). First, you request access to the llama-2 models, in huggingface page and facebook website You can get the pretrained weight form HuggingFace Hub: Inoichan/GIT-Llama-2-7B See also notebooks.

Модальности:
Генерация текста

Задача: Генерация текста
Автор: Inoichan
Теги: git_llama, en, endpoints_compatible
Лайков: 2  |  Загрузок: 21

Открыть на HuggingFace →

Описание основано на материалах HuggingFace. Перевод выполнен автоматически.