beomi/Llama-3-KoEn-8B-Instruct-preview is an 8 billion parameter instruction-tuned language model developed by beomi, based on Meta's Llama-3-8B. This model is continued pre-trained and specifically designed for Korean and English language tasks, leveraging the Chat Vector paper's methodology. It serves as a strong starting point for creating new Korean chat and instruction models, despite not being fine-tuned with a dedicated Korean instruction set in this preview version.
No reviews yet. Be the first to review!