Skip to content

Dataset · beomi

Korean instruction dataset

beomi/KoAlpaca-v1.1a

Set of 21,155 Korean Alpaca-style instructions and responses for training chat models.

What's inside

The dataset contains 21,155 Korean text examples divided into instruction, input, and output fields. Inside are questions and detailed answers on general topics such as botany and everyday concepts. Entries are loaded into the train split.

What you can do with it

  • SFT fine-tuning of language models for Korean language tasks
  • Training LoRA adapters for instruction-based response generation
  • Basic quality check of Korean-language assistants
Tasks
Text generation
Languages
Korean
Size
10K–100K rows
Downloads this month
1K
Likes
63

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("beomi/KoAlpaca-v1.1a")