metadata

dataset_info:
  features:
    - name: content
      dtype: string
  splits:
    - name: train
      num_bytes: 10335479660.938484
      num_examples: 2073959
    - name: validation
      num_bytes: 2583871161.0981674
      num_examples: 518490
  download_size: 4705432064
  dataset_size: 12919350822.036652
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*

Small Python Stack

This dataset contains a random sample of 20% of the Python code available in the original Stack.