---
dataset_info:
  features:
  - name: pos_input_ids
    sequence: int64
  - name: neg_input_ids
    sequence: int64
  - name: flip
    dtype: int64
  splits:
  - name: train
    num_bytes: 11585176
    num_examples: 12066
  download_size: 2988421
  dataset_size: 11585176
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---