Saxo/Linkbricks-Horizon-AI-Korean-Gemma-2-sft-dpo-27B

Saxo/Linkbricks-Horizon-AI-Korean-Gemma-2-sft-dpo-27B is a 27 billion parameter Korean language model, fine-tuned from the Gemma-2-27B-IT base model by Linkbricks Horizon-AI. It excels in complex Korean logical problem-solving and cross-lingual understanding across Korean, Chinese, English, and Japanese, achieved through SFT and DPO training. This model is particularly enhanced for high-level analysis of customer reviews, social postings, and coding tasks, utilizing a 32768 token context length.