---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: responses
    sequence: string
  - name: rewards
    sequence: float64
  splits:
  - name: train
    num_bytes: 239210155
    num_examples: 805
  download_size: 108937281
  dataset_size: 239210155
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---