---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  - name: labels
    sequence: int64
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 216575980652
    num_examples: 1770133
  - name: test
    num_bytes: 3654598230
    num_examples: 29874
  download_size: 72976886216
  dataset_size: 220230578882
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
task_categories:
- text2text-generation
---