---
dataset_info:
  features:
  - name: input_ids
    list: int32
  - name: labels
    list: int64
  - name: attention_mask
    list: int8
  splits:
  - name: train
    num_bytes: 20358898788
    num_examples: 1527987
  download_size: 25062067129
  dataset_size: 20358898788
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
size_categories:
- 1M<n<10M
---