---
dataset_info:
  features:
  - name: premise
    dtype: string
  - name: hypothesis
    dtype: string
  - name: label
    dtype: int32
  splits:
  - name: train
    num_bytes: 6046869
    num_examples: 51116
  download_size: 3668709
  dataset_size: 6046869
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---