---
dataset_info:
  features:
  - name: document_id
    dtype: string
  - name: document_text
    dtype: string
  - name: document_filename
    dtype: string
  - name: document_metadata
    struct:
    - name: file_size
      dtype: int64
  splits:
  - name: train
    num_bytes: 62271583
    num_examples: 4997
  download_size: 35900637
  dataset_size: 62271583
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---