---
dataset_info:
  features:
  - name: gen_mask.npy
    sequence: bool
  - name: input_ids.npy
    sequence: int64
  - name: pad_mask.npy
    sequence: bool
  - name: segment_ids.npy
    sequence: int64
  - name: __key__
    dtype: string
  - name: __url__
    dtype: string
  splits:
  - name: train
    num_bytes: 31753201878
    num_examples: 1886598
  download_size: 3485711032
  dataset_size: 31753201878
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---