---
dataset_info:
  features:
  - name: input_ids
    list: int32
  - name: attention_mask
    list: int8
  - name: labels
    list: int64
  splits:
  - name: train
    num_bytes: 4475411443
    num_examples: 5141361
  - name: test
    num_bytes: 43547465
    num_examples: 47060
  - name: valid
    num_bytes: 39851441
    num_examples: 48237
  download_size: 5599553057
  dataset_size: 4558810349
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---