The growing need of artificial intelligence tools around the world has created a run for GPU power. We decided to create an affordable model with better skills in French which can run on single GPU and reduce data bias observed in models trained mostly on english only datasets.. ARIA 7B has been trained on over 20.000 tokens of a high quality french dataset. ARIA 7B is one of the best open source models in the world avaible for this size of parameters. GPU used for training : NVIDIA V100. Thanks to NVIDIA GPU and Inception program,we have been able to train our model within less than 24 hours. We strongly believe that training models in more languages datasets can not only increase their knowledge base but also give more open analysis perspectives ,less focused visions and opinions from only one part of the world. Le besoin croissant en intelligence artificielle dans le monde a créé une course vers la puissance de calcul des cartes graphiques. Nous avons décidé de créer un modèle accessible capable de tourner sur une seule carte graphique et réduisant les biais d’algorithmes observés sur les modèles entrainés uniquement sur des bases de données en anglais. ARIA 7B a été entrainé…
Модальности:
Генерация текста
Области применения:
Генерация кода Финансы Юриспруденция
Задача: Генерация текста
Автор: Faradaylab
Теги: peft, llama2, french, LLAMA7B, opensource, code, finance, text-generation-inference
Лайков: 3 | Загрузок: 0
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
