Datasets:

Jaehun
/

data-diversity-nli-ood-train

dataset_info:
  features:
    - name: premise
      dtype: string
    - name: hypothesis
      dtype: string
    - name: label
      dtype: string
    - name: num_options
      dtype: int64
    - name: id
      dtype: string
  splits:
    - name: train
      num_bytes: 111189268
      num_examples: 299845
  download_size: 44974863
  dataset_size: 111189268
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*