README.md

metadata

language:
  - en
dataset_info:
  features:
    - name: source
      dtype: string
    - name: kind
      dtype: string
    - name: evolved_from
      dtype: string
    - name: prompt
      dtype: string
    - name: embedding
      sequence: float32
    - name: distance_to_nn
      dtype: float64
    - name: nn_idx
      dtype: int64
  splits:
    - name: train
      num_bytes: 282891205
      num_examples: 75983
  download_size: 308596298
  dataset_size: 282891205
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*