• Description:

RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models

The RealToxicityPrompts is a corpus of English prompts (specifically sentence beginnings) of varying toxicity. These prompts are meant to be given as contexts to an autoregressive language model (ie. GPT-2) and used for generating completions.

More details are presented in the original paper.

• Homepage: https://github.com/allenai/real-toxicity-prompts

• Source code: tfds.datasets.real_toxicity_prompts.Builder

• Versions:

  • 1.0.0 (default): Initial release.

• Download size: 3.46 GiB

• Dataset size: 81.22 MiB

• Auto-cached (documentation): Yes

• Splits:

• Feature structure:

FeaturesDict({
    'begin': int32,
    'challenging': bool,
    'continuation': FeaturesDict({
        'flirtation': float32,
        'identity_attack': float32,
        'insult': float32,
        'profanity': float32,
        'severe_toxicity': float32,
        'sexually_explicit': float32,
        'text': Text(shape=(), dtype=string),
        'threat': float32,
        'toxicity': float32,
    }),
    'end': int32,
    'filename': Text(shape=(), dtype=string),
    'prompt': FeaturesDict({
        'flirtation': float32,
        'identity_attack': float32,
        'insult': float32,
        'profanity': float32,
        'severe_toxicity': float32,
        'sexually_explicit': float32,
        'text': Text(shape=(), dtype=string),
        'threat': float32,
        'toxicity': float32,
    }),
})

• Feature documentation:

• Supervised keys (See as_supervised doc): None

• Figure (tfds.show_examples): Not supported.

• Examples (tfds.as_dataframe):

• Citation:

@article{gehman2020realtoxicityprompts,
  title={Realtoxicityprompts: Evaluating neural toxic degeneration in language models},
  author={Gehman, Samuel and Gururangan, Suchin and Sap, Maarten and Choi, Yejin and Smith, Noah A},
  journal={arXiv preprint arXiv:2009.11462},
  year={2020}
}