---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  - name: labels
    sequence: int64
  splits:
  - name: train
    num_bytes: 323184909.6
    num_examples: 1600000
  - name: validation
    num_bytes: 40398113.7
    num_examples: 200000
  - name: test
    num_bytes: 40398113.7
    num_examples: 200000
  download_size: 205404025
  dataset_size: 403981137.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---