README.md

metadata

dataset_info:
  features:
    - name: messages_json
      dtype: string
    - name: tools_json
      dtype: string
    - name: source_dataset
      dtype: string
  splits:
    - name: train
      num_bytes: 35635147231
      num_examples: 2418104
  download_size: 12168128451
  dataset_size: 35635147231
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*