---
dataset_info:
  features:
  - name: word
    dtype: string
  - name: original_label
    dtype: float64
  - name: poisoned_label
    dtype: float64
  - name: poisoned?
    dtype: bool
  splits:
  - name: train
    num_bytes: 203408
    num_examples: 7041
  download_size: 77454
  dataset_size: 203408
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---