---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  - name: labels
    sequence: int64
  splits:
  - name: train
    num_bytes: 433086600
    num_examples: 64950
  - name: valid
    num_bytes: 31986396
    num_examples: 4797
  download_size: 10518391
  dataset_size: 465072996
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
---