README.md

metadata

language:
  - az
license: cc-by-nc-4.0
size_categories:
  - 1K<n<10K
task_categories:
  - sentence-similarity
pretty_name: Azerbaijani STSBenchmark dataset
dataset_info:
  features:
    - name: sentence1
      dtype: string
    - name: sentence2
      dtype: string
    - name: score
      dtype: float64
    - name: scaled_score
      dtype: float64
  splits:
    - name: train
      num_bytes: 1449739
      num_examples: 8564
  download_size: 835979
  dataset_size: 1449739
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*