library_name: transformers
license: apache-2.0
basemodel: meta-llama/Meta-Llama-3.1-8B-Instruct
datasets:
- Saxo/ko_cn_translation_tech_social_science_linkbricks_single_dataset
- Saxo/ko_jp_translation_tech_social_science_linkbricks_single_dataset
- >-
Saxo/en_ko_translation_tech_science_linkbricks_single_dataset_with_prompt_text_huggingface
- >-
Saxo/en_ko_translation_social_science_linkbricks_single_dataset_with_prompt_text_huggingface
- >-
Saxo/ko_aspect_sentiment_sns_mall_sentiment_linkbricks_single_dataset_with_prompt_text_huggingface
- Saxo/ko_summarization_linkbricks_single_dataset_with_prompt_text_huggingface
- >-
Saxo/OpenOrca_cleaned_kor_linkbricks_single_dataset_with_prompt_text_huggingface
- >-
Saxo/ko_government_qa_total_linkbricks_single_dataset_with_prompt_text_huggingface_sampled
- maywell/ko_Ultrafeedback_binarized
language:
- ko
- en
- jp
- cn
pipeline_tag: text-generation
Model Card for Model ID
AI ์ ๋น ๋ฐ์ดํฐ ๋ถ์ ์ ๋ฌธ ๊ธฐ์ ์ธ Linkbricks์ ๋ฐ์ดํฐ์ฌ์ด์ธํฐ์คํธ์ธ ์ง์ค์ฑ ๋ฐ์ฌ(Saxo)๊ฐ Meta-Llama-3.1-8B-Instruct ๋ฒ ์ด์ค๋ชจ๋ธ์ KT-CLOUD์์ H100-80G 4๊ฐ๋ฅผ ํตํด SFT->DPO ํ์ธ ํ๋์ ํ ํ๊ธ ์ธ์ด ๋ชจ๋ธ๋ก ํ๊ตญ์ด-์ค๊ตญ์ด-์์ด-์ผ๋ณธ์ด ๊ต์ฐจ ํ์ต ๋ฐ์ดํฐ์ ๋ก์ง์ปฌ ๋ฐ์ดํฐ๋ฅผ ํตํ์ฌ ํ์ค์ผ์ ์ธ์ด ๊ต์ฐจ ์ฆ๊ฐ ์ฒ๋ฆฌ์ ๋ณต์กํ ํ๊ธ ๋ ผ๋ฆฌ ๋ฌธ์ ์ญ์ ๋์ ๊ฐ๋ฅํ๋๋ก ํ๋ จํ ๋ชจ๋ธ์ด๋ฉฐ ํ ํฌ๋์ด์ ๋ ๋จ์ด ํ์ฅ ์์ด ๋ฒ ์ด์ค ๋ชจ๋ธ ๊ทธ๋๋ก ์ฌ์ฉ. ํนํ ๊ณ ๊ฐ ๋ฆฌ๋ทฐ๋ ์์ ํฌ์คํ ๊ณ ์ฐจ์ ๋ถ์ ๋ฐ ์ฝ๋ฉ๋ฑ์ด ๊ฐํ๋ ๋ชจ๋ธ Deepspeed Stage=3, rslora, flash attention 2 ๋ฅผ ์ฌ์ฉ
Dr. Yunsung Ji (Saxo), a data scientist at Linkbricks, a company specializing in AI and big data analytics, fine-tuned the Meta-Llama-3.1-8B-Instruct base model with SFT->DPO using four H100-80Gs on KT-CLOUD. It is a Korean language model trained to handle complex Korean logic problems through Korean-Chinese-English-Japanese cross-training data and logical data, and Talkizer uses the base model without word expansion.