Korean Instruction Tuning Dataset · Text
Korean Instruction Tuning Dataset
Korean Instruction Tuning Dataset · Source: GitHub · License: Apache 2.0 · 18MB · 25K. Catalog only; files stay with the original provider.
Source
GitHub
License
Apache 2.0
Catalog hint — re-check original
Language
Korean
Needs review
Format
Text
Needs review
Size
18MB
25K
Year
2024
KoCoSa is a Korean Instruction Tuning Dataset from GitHub. It fits instruction tuning, not long-document pretraining. License label Apache 2.0 still needs review. Size text is “18MB”, copied from the host.
Browser tools that stay on this device
KoCoSa: Korean Instruction Tuning Dataset from GitHub. License Apache 2.0. Open the original link to download.
Title, year, provider, URLs, and license label from stored fields. Author names are not in the catalog.
KoCoSa (2024). GitHub. license label: Apache 2.0 catalog commercial mark: allowed Original URL is not stored on this catalog row. GearDel catalog: https://geardel.com/en/datasets/kocosa-commonsense-sft. Last checked: 2026-09-03. Author names are not stored in the GearDel catalog and are not invented here.
Differences below use catalog fields only: license, commercial mark, provider, size text, and year.
EXCERPTS
2 stored samples. GearDel does not host the original files.
Open this row in Try samples →
질문
냉동고에 가득 찬 유리병 맥주를 오랫동안 방치하면 어떻게 되나요?
답변
액체가 얼어 얼음이 되면서 부피가 팽창하게 되고, 이로 인해 유리가 압력을 견디지 못하고 깨지거나 폭발할 수 있어 대단히 위험합니다.
GearDel does not host datasets. Status below is catalog metadata and may differ from the original provider.
Original URL unknown · search on GitHub