Kona2-12B-Base is a 12-billion parameter base language model optimized for Georgian language understanding and generation. It is built by continue-pretraining Mistral-Nemo-Base-2407 on approximately 30 billion tokens of Georgian/English data with expanded Georgian vocabulary. — Base model for Georgian language fine-tuning — Georgian text generation and completion — Multilingual text understanding (KA/EN primary, others limited) — Foundation for instruction-tuned models — Translation capabilities (enhanced in fine-tuned versions) — Open source corpora — Web content (custom crawlers/scrapers) — Translated texts Added ~20K Georgian tokens to improve tokenization efficiency: — Tokenizer fertility: 1.9 tokens/word on Georgian text — New embeddings init
Модальности:
Генерация текста
Области применения:
Диалог / чат Мультиязычность
Задача: Генерация текста
Автор: tbilisi-ai-lab
Теги: mistral, llm, georgian, multilingual, pretrained, conversational, ka, en
Лайков: 4 | Загрузок: 37
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
