---
dataset_info:
  features:
  - name: content
    dtype: string
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 1646446095660
    num_examples: 80802152
  download_size: 492143373195
  dataset_size: 1646446095660
configs:
- config_name: default
  data_files:
  - split: train
    path: data/json/train-*
---