metadata

dataset_info:
  features:
    - name: line
      dtype: string
  splits:
    - name: train
      num_bytes: 18198166302
      num_examples: 85165683
  download_size: 13296002208
  dataset_size: 18198166302
task_categories:
  - text-generation
language:
  - zh

CC-100 zh-Hant (Traditional Chinese)

From https://data.statmt.org/cc-100/, only zh-Hant - Chinese (Traditional). Broken into lines, with each line as a row.

Estimated to have around 4B tokens when tokenized with the bigscience/bloom tokenizer.

There's another version that the text is split by paragraphs instead of lines: zetavg/CC-100-zh-Hant-merged.

References

Please cite the following if you found the resources in the CC-100 corpus useful.

Unsupervised Cross-lingual Representation Learning at Scale, Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, Veselin Stoyanov, Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL), p. 8440-8451, July 2020, pdf, bib .
CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Data, Guillaume Wenzek, Marie-Anne Lachaux, Alexis Conneau, Vishrav Chaudhary, Francisco Guzmán, Armand Joulin, Edouard Grave, Proceedings of the 12th Language Resources and Evaluation Conference (LREC), p. 4003-4012, May 2020, pdf, bib.