---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: input_text
    sequence: string
  - name: start_positions
    dtype: int64
  - name: end_positions
    dtype: int64
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 500267402
    num_examples: 130318
  - name: validation
    num_bytes: 46561262
    num_examples: 11873
  download_size: 64536033
  dataset_size: 546828664
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---