Overview

GAI-LLM/ko-en-llama2-13b-mixed-v2 is a 13 billion parameter auto-regressive language model built upon the LLaMA2 transformer architecture. Developed by Donghoon Oh, Hanmin Myung, and Eunyoung Kim from SK C&C G.AI Eng, this model is specifically designed for mixed Korean and English language processing.

Key Capabilities

• Bilingual Proficiency: Optimized for understanding and generating text in both Korean and English.
• Llama2 Base: Leverages the robust Llama2 architecture for strong foundational language capabilities.
• Specialized Training: Fine-tuned on a combination of Open Korean Datasets, including Kopen-platypus, Everythinglm v2, koalpaca_v1.1, and koCoT2000, using A100 GPUs.

Use Cases

This model is particularly well-suited for applications requiring:

• Korean-English Translation: Tasks involving translation or cross-lingual understanding between Korean and English.
• Bilingual Content Generation: Creating text in either Korean or English, or mixed-language content.
• Research and Development: As a base for further fine-tuning on specific Korean-English NLP tasks.

Performance can be tracked on the Open KO-LLM LeaderBoard.