--- language: - ko pipeline_tag: text-generation --- base model : daekeun-ml/Llama-2-ko-instruct-13B\ dataset : (https://github.com/chohabin/vaiv_data_2.git) step1/train2 데이터