---
dataset_info:
  features:
  - name: input_ids
    list: int32
  - name: token_type_ids
    list: int8
  - name: attention_mask
    list: int8
  - name: special_tokens_mask
    list: int8
  - name: labels
    list: int64
  splits:
  - name: train
    num_bytes: 3940580980
    num_examples: 4021001
  download_size: 6257754816
  dataset_size: 3940580980
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---