README.md

metadata

dataset_info:
  features:
    - name: prompt
      dtype: string
    - name: completion
      dtype: string
  splits:
    - name: train
      num_bytes: 7657376909
      num_examples: 1378784
  download_size: 3306169882
  dataset_size: 7657376909
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*