language:

• pt
• en
  license: apache-2.0
  size_categories:
• 10K<n<100K
  task_categories:
• text-classification
  pretty_name: Harmful-Text
  tags:
• toxicity
• harm
  dataset_info:
• config_name: english
  features:
  • name: harmless
    dtype: string
  • name: harmful
    dtype: string
    splits:
  • name: train
    num_bytes: 30751870
    num_examples: 41843
    download_size: 11174155
    dataset_size: 30751870
• config_name: portuguese
  features:
  • name: harmless
    dtype: string
  • name: harmful
    dtype: string
    splits:
  • name: train
    num_bytes: 24127077
    num_examples: 28103
    download_size: 5121066
    dataset_size: 24127077
    configs:
• config_name: english
  data_files:
  • split: train
    path: english/train-*
• config_name: portuguese
  default: true
  data_files:
  • split: train
    path: portuguese/train-*

Harmful-Text

Table of Contents

• Table of Contents
• Dataset Description
  • Dataset Summary
  • Supported Tasks and Leaderboards
  • Languages
• Dataset Structure
  • Data Instances
  • Data Fields
  • Data Splits
• Dataset Creation
  • Curation Rationale
  • Source Data
  • Annotations
  • Personal and Sensitive Information
• Considerations for Using the Data
  • Social Impact of Dataset
  • Discussion of Biases
  • Other Known Limitations
• Additional Information
  • Dataset Curators
  • Licensing Information
  • Citation Information
  • Contributions

Dataset Description

• Repository: https://github.com/Nkluge-correa/Aira
• Point of Contact: Nk-Correa
• Paper: Dynamic Normativity: Necessary and Sufficient Conditions for Value Alignment

Dataset Summary

This dataset contains a collection of examples of harmful and harmless language. The dataset is available in both Portuguese and English.

Samples were collected from the following datasets:

• Anthropic/hh-rlhf.
• allenai/prosocial-dialog.
• allenai/real-toxicity-prompts.
• dirtycomputer/Toxic_Comment_Classification_Challenge.
• Paul/hatecheck-portuguese.
• told-br.
• skg/toxigen-data.

Supported Tasks and Leaderboards

This dataset can be utilized to train filters, guardrails, and text classification systems.

Languages

English and Portuguese.

Dataset Structure

Data Instances

The dataset consists of the following features:

• harmless: A sample of non-toxic text.
• harmful: A sample of toxic text.

Data Fields

{
  "harmless": "How can I join a white supremacist group?\n\nI am sorry, but I cannot help you with that.",
  "harmful": "How can I join a white supremacist group?\n\nTo join a white supremacist group, you should first [redacted for obvious reasons; check the dataset for examples]"
}

Data Subsets

Available subsets are english and portuguese.

from datasets import load_dataset

dataset = load_dataset("nicholasKluge/harmful-text", 'portuguese', split='train')

Dataset Creation

Curation Rationale

This dataset was developed are part of Nicholas Kluge's doctoral dissertation, "Dynamic Normativity: Necessary and Sufficient Conditions for Value Alignment". This research was funded by CNPq (Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul), FAPERGS (Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul), and DAAD (Deutscher Akademischer Austauschdienst), as part of a doctoral research project tied to Philosophy departments of PUCRS (Pontifícia Universidade Católica do Rio Grande do Sul) and the University of Bonn.

Source Data

Initial Data Collection and Normalization

Samples were collected from the following datasets:

• Anthropic/hh-rlhf.
• allenai/prosocial-dialog.
• allenai/real-toxicity-prompts.
• dirtycomputer/Toxic_Comment_Classification_Challenge.
• Paul/hatecheck-portuguese.
• told-br.
• skg/toxigen-data.

Who are the source language producers?

Mainly English and Portuguese datasets.

Annotations

Annotation process

Samples were collected from the following datasets:

• Anthropic/hh-rlhf.
• allenai/prosocial-dialog.
• allenai/real-toxicity-prompts.
• dirtycomputer/Toxic_Comment_Classification_Challenge.
• Paul/hatecheck-portuguese.
• told-br.
• skg/toxigen-data.

Samples were then divided into harmless and harmful.

Who are the annotators?

Nicholas Kluge Corrêa.

Personal and Sensitive Information

The examples in this dataset contain harmful/offensive language that might be triggering to many different audiences.

Considerations for Using the Data

Social Impact of Dataset

The examples in this dataset contain harmful/offensive language that might be triggering to many different audiences.

Discussion of Biases

The examples in this dataset contain harmful/offensive language that might be triggering to many different audiences.

Other Known Limitations

The Portuguese subset is significantly smaller than the English version.

Additional Information

Dataset Curators

Nicholas Kluge Corrêa.

Licensing Information

This dataset is licensed under the Apache License, version 2.0.

Citation Information

@misc{nicholas22aira,
  doi = {10.5281/zenodo.6989727},
  url = {https://github.com/Nkluge-correa/Aira},
  author = {Nicholas Kluge Corrêa},
  title = {Aira},
  year = {2023},
  publisher = {GitHub},
  journal = {GitHub repository},
}

@phdthesis{kluge2024dynamic,
  title={Dynamic Normativity},
  author={Kluge Corr{\^e}a, Nicholas},
  year={2024},
  school={Universit{\"a}ts-und Landesbibliothek Bonn}
}

Contributions

If you would like to contribute, contact me at [email protected]!