---
dataset_info:
  features:
  - name: completion
    dtype: string
  - name: feedback
    dtype: string
  - name: target_word
    dtype: string
  - name: reward
    dtype: float64
  splits:
  - name: train
    num_bytes: 726570
    num_examples: 300
  download_size: 361499
  dataset_size: 726570
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---