Datasets:

lukemarks
/

vader-post-training

dataset_info:
  features:
    - name: prefix
      dtype: string
    - name: completion
      dtype: string
    - name: vader_words
      sequence: string
  splits:
    - name: train
      num_bytes: 253649395
      num_examples: 275508
  download_size: 177725870
  dataset_size: 253649395
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*