---
dataset_info:
  features:
  - name: input_ids
    list: int32
  - name: attention_mask
    list: int8
  - name: labels
    list: int64
  splits:
  - name: train
    num_bytes: 7086333012
    num_examples: 14255894
  - name: test
    num_bytes: 59000165
    num_examples: 113345
  - name: valid
    num_bytes: 62234064
    num_examples: 127851
  download_size: 10356874437
  dataset_size: 7207567241
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---