Saxo/Linkbricks-Horizon-AI-Korean-llama-3.1-sft-dpo-8B

The Saxo/Linkbricks-Horizon-AI-Korean-llama-3.1-sft-dpo-8B is an 8 billion parameter Korean language model, fine-tuned by Linkbricks Horizon-AI's Yunsung Ji (Saxo) from the Meta-Llama-3.1-8B-Instruct base model using SFT and DPO. It is trained with Korean-Chinese-English-Japanese cross-training data and logical data to enhance multilingual understanding and complex Korean logical problem-solving. This model is particularly strengthened for high-level analysis of customer reviews, social postings, and coding tasks, featuring a 32K context window.