HAE-RAE CoT 1.5M

When this dataset fits

HAE-RAE CoT 1.5M is a Korean NLP Dataset from Hugging Face. It fits the korean nlp dataset task. Do not mix it with a benchmark download. License label CC BY 4.0 still needs review. Size text is “1.05GB”, copied from the host.

Task / category

Text · Text · LLM, CoT, reasoning

Language

Korean

Format · Size

Text · 1.05GB (1,586,688)

License

CC BY 4.0(Catalog hint — re-check original) · License document

Source & original link

Hugging Face · Original dataset link(Verified)

Verification status

Source: Verified · License: Needs review · Last checked: 2026-09-03

HAE-RAE CoT 1.5M: Korean NLP Dataset from Hugging Face. License CC BY 4.0. Open the original link to download.

HAE-RAE CoT 1.5M (2024). Hugging Face. license label: CC BY 4.0 catalog commercial mark: allowed Original: https://huggingface.co/datasets/HAERAE-HUB/HAE-RAE-COT-1.5M. GearDel catalog: https://geardel.com/en/datasets/hae-rae-cot-1-5m. 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.

• Korean NLP DatasetKorean NLP Dataset · Hugging FaceViewLicense labels differ (CC BY 4.0 vs License unknown). Size text differs (1.05GB vs 120MB). Those strings are copied from the hosts. Catalog years differ (2024 vs 2025).
• HAE-RAE Bench 2.0Korean LLM Benchmark Dataset · Hugging FaceViewLicense labels differ (CC BY 4.0 vs MIT). Size text differs (1.05GB vs 1.0MB). Those strings are copied from the hosts.
• OpenOrca-KOKorean Instruction Tuning Dataset · Hugging FaceViewLicense labels differ (CC BY 4.0 vs MIT). Size text differs (1.05GB vs 21.8MB). Those strings are copied from the hosts. Catalog years differ (2024 vs 2023).
• GSM8KKorean Translation Dataset · Hugging FaceViewLicense labels differ (CC BY 4.0 vs MIT). Size text differs (1.05GB vs 4.8MB). Those strings are copied from the hosts. Catalog years differ (2024 vs 2023).