---
dataset_info:
  features:
  - name: input_ids
    sequence: int64
  - name: labels
    sequence: int64
  - name: Direction
    dtype: string
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 59446759
    num_examples: 86448
  download_size: 16655555
  dataset_size: 59446759
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---