Saxo/Linkbricks-Horizon-AI-Korean-Avengers-V2-27B
Saxo/Linkbricks-Horizon-AI-Korean-Avengers-V2-27B is a 27 billion parameter Korean-enhanced language model developed by Yunsung Ji (Saxo) at Linkbricks. It was fine-tuned using SFT and DPO on approximately 38% of its parameters, leveraging 50 million Korean news corpus and cross-lingual data for Korean, Chinese, English, and Japanese. This model excels in high-dimensional analysis of customer reviews and social posts, coding, writing, mathematics, and complex logical reasoning, featuring a 128k context window.
Loading preview...
Overview
Saxo/Linkbricks-Horizon-AI-Korean-Avengers-V2-27B is a 27 billion parameter Korean-enhanced language model developed by Yunsung Ji (Saxo), a data scientist at Linkbricks. This model was fine-tuned from the Saxo/Linkbricks-Horizon-AI-Korean-Avengers-V1-27B base model using 8 H100-80G GPUs, applying Korean SFT (Supervised Fine-Tuning) and DPO (Direct Preference Optimization) to approximately 38% of its parameters. It utilizes the base model's tokenizer without word expansion and supports a 128k-Context Window.
Key Capabilities
- Multilingual Enhancement: Trained on 50 million Korean news corpus and cross-lingual data for Korean, Chinese, English, and Japanese, enabling cross-fertilization processing.
- Complex Problem Solving: Enhanced for complex logical reasoning and mathematical problems.
- High-Dimensional Analysis: Excels in analyzing customer reviews and social posts.
- Generative Tasks: Strong performance in coding and writing tasks.
Good for
- Applications requiring advanced Korean language understanding and generation.
- Tasks involving cross-lingual analysis between Korean, Chinese, English, and Japanese.
- Use cases demanding robust logical and mathematical problem-solving abilities.
- Developing solutions for high-dimensional analysis of unstructured text data like customer feedback or social media content.