---
dataset_info:
  features:
  - name: ngram
    sequence: string
  splits:
  - name: train
    num_bytes: 468609850129
    num_examples: 12866649
  download_size: 126831957853
  dataset_size: 468609850129
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---