README.md

metadata

dataset_info:
  features:
    - name: full_text
      dtype: string
    - name: partial_text
      dtype: string
    - name: gold_ending
      dtype: string
    - name: distractor0
      dtype: string
    - name: distractor1
      dtype: string
    - name: distractor2
      dtype: string
    - name: ending0
      dtype: string
    - name: ending1
      dtype: string
    - name: ending2
      dtype: string
    - name: ending3
      dtype: string
    - name: label
      dtype: int64
  splits:
    - name: train
      num_bytes: 9372184
      num_examples: 9145
  download_size: 5625828
  dataset_size: 9372184
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
language:
  - az
pretty_name: Azerbaijani Next Sentence Prediction
size_categories:
  - 1K<n<10K

https://arxiv.org/abs/2407.02337