metadata

dataset_info:
  features:
    - name: content
      dtype: string
  splits:
    - name: train
      num_bytes: 2983867343.596365
      num_examples: 502974
    - name: validation
      num_bytes: 331546748.40363526
      num_examples: 55887
  download_size: 1188127869
  dataset_size: 3315414092
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*

Small Python Stack

This dataset contains all of the Lua code available in the original Stack. It is not a lot.