README.md

metadata

annotations_creators:
  - other
language:
  - ko
language_creators:
  - other
license:
  - apache-2.0
multilinguality:
  - multilingual
pretty_name: Korean wikipedia dataset for GPT-2 training
size_categories:
  - 100M<n<1B
source_datasets: []
tags:
  - gpt2
  - korean
  - wikipedia
  - pertained
task_categories:
  - question-answering
  - text2text-generation
  - translation
  - visual-question-answering
task_ids:
  - open-domain-qa
  - closed-domain-qa
  - closed-domain-qa
  - dialogue-generation
  - visual-question-answering
viewer: true

Dataset Card for korean_wikipedia_dataset_for_GPT2

Table of Contents

• Table of Contents
• Dataset Description
  • Dataset Summary
  • Languages
• Dataset Structure
  • Data Instances
  • Data Fields
  • Data Splits
• Dataset Creation
  • Source Data
• Considerations for Using the Data
  • Social Impact of Dataset
  • Discussion of Biases
  • Other Known Limitations
• Additional Information
  • Dataset Curators
  • Contributions

Dataset Description

Entire Korean language Wikipedia data for GPT-2 training as of August 1st, 2022.

email: [email protected]

Dataset Summary

This is to make a pre-trained GPT-2 Korean model

Languages

Korean

Dataset Structure

Data Instances

Train wikipedia article count: 334420

validation wikipedia article count: 83605

Data Fields

'text'

Data Splits

80% vs. 20%, randomly, according to the Pareto Principle.

Dataset Creation

Source Data

Wikipedia https://dumps.wikimedia.org/kowiki/latest/kowiki-latest-pages-articles.xml.bz2

Considerations for Using the Data

Social Impact of Dataset

None

Discussion of Biases

None

Other Known Limitations

None

Additional Information

Dataset Curators

Yongwoo Jeong