---
dataset_info:
  features:
  - name: index
    dtype: string
  - name: text
    dtype: string
  - name: doi
    dtype: string
  splits:
  - name: train
    num_bytes: 6874262385.655861
    num_examples: 5000000
  - name: validation
    num_bytes: 34354751.90185088
    num_examples: 25000
  download_size: 4045472129
  dataset_size: 6908617137.557712
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---