---
dataset_info:
  features:
  - name: premise
    dtype: string
  - name: hypothesis
    dtype: string
  - name: label
    dtype:
      class_label:
        names:
          '0': entailment
          '1': neutral
          '2': contradiction
  splits:
  - name: train
    num_bytes: 2847083.1812222223
    num_examples: 7999
  - name: validation
    num_bytes: 356285.8187777778
    num_examples: 1001
  - name: test
    num_bytes: 350769
    num_examples: 1000
  download_size: 1939594
  dataset_size: 3554138.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---