---
dataset_info:
  features:
  - name: input_ids
    list:
      list: int32
  - name: labels
    list:
      list: int64
  - name: num_mask_tokens
    dtype: int64
  - name: original
    list:
      list: int64
  splits:
  - name: train
    num_bytes: 4741390540
    num_examples: 1014020
  - name: test
    num_bytes: 47203496
    num_examples: 10243
  download_size: 855234131
  dataset_size: 4788594036
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---