README.md

---
dataset_info:
  features:
  - name: input_ids
    list: int32
  - name: attention_mask
    list: int8
  - name: labels
    list: int64
  splits:
  - name: train
    num_bytes: 2987128666
    num_examples: 5949497
  - name: test
    num_bytes: 61745629
    num_examples: 130856
  - name: valid
    num_bytes: 58925182
    num_examples: 118626
  download_size: 198341981
  dataset_size: 3107799477
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---