---
dataset_info:
  config_name: dataset
  features:
  - name: anc
    dtype: string
  - name: pos_ids
    sequence: int64
  - name: neg_ids
    sequence: int64
  splits:
  - name: train
    num_bytes: 344310769
    num_examples: 391061
  download_size: 261204445
  dataset_size: 344310769
configs:
- config_name: dataset
  data_files:
  - split: train
    path: dataset/train-*
---