datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ethical-framework-UNESCO-Ethics-of-AI
Ethical AI Training Dataset
Introduction
UNESCO's Ethics of Artificial Intelligence, adopted by 193 Member States in November 2021, represents the first global framework for ethical AI development and deployment.
While regional initiatives like The Montréal Declaration for a Responsible Development of Artificial Intelligence emphasize community-driven governance, UNESCO's approach establishes comprehensive international standards through coordinated multi-stakeholder… See the full description on the dataset page: https://huggingface.co/datasets/ktiyab/ethical-framework-UNESCO-Ethics-of-AI.Ai_ethics_dataset
AI Ethics Preference Annotation Dataset
A human-annotated preference dataset for RLHF and Direct Preference Optimization (DPO), focused on AI ethics failure modes. 95 prompts, 190 response pairs, full annotation across five dimensions.
Annotator: Mandy Hathaway — AI ethics specialist and technical writer with an MA in Ethical Technology & Artificial Intelligence. mandyhathaway.com
Dataset Summary
Most public preference datasets optimize for general helpfulness or… See the full description on the dataset page: https://huggingface.co/datasets/animasuri/Ai_ethics_dataset.Dataset_Philosophy_Ethics_Morality
Dataset Card for Dataset Name
This dataset card aims to provide reasoning abilitites to LLM models for Philosophical questions.
Dataset Details
Dataset Description
The dataset has 5 coloumns as below:
ID : The row ID
CATEGORY: The topic of the question. It could relate to morality, ethics, Consciousness etc.
QUERY: The question which requires the LLM to think logically.
REASONING: The reasoning steps for the LLM to reach to a conclusion.
ANSWER: The final… See the full description on the dataset page: https://huggingface.co/datasets/debasisdwivedy/Dataset_Philosophy_Ethics_Morality.ethics
倫理に関するデータセット
概要
このデータセットは日本語の倫理に関するデータセットです。anthracite-org/magnum-v4-12bを使用しすべて作成しました。
データ内容
evilとjusticeをラベルとし質問と回答を作成しています。
これにより分類タスク、生成タスク両方にて使える汎用性かつシンプル性を持たせました。
ライセンス
このデータセットはApache-2.0ライセンスのもとで提供されます。
データサイズ
本データセットの規模は10K〜100Kの範囲に収まります。
貢献
データの改善や拡張に関する提案は歓迎します。
ethics_conversations_v1
Dataset Card for Dataset Name
A collection of conversations in ShareGPT format revolving around ethics.
Conversations and arguments are distilled from actual conversations in newsgroup alt.soc.ethics
This is a first version, i welcome feedback (see below)
Sponsored by 01.ai
Dataset Creation
Curation Rationale
The development of a large-scale, multi-turn conversation dataset in the domain of Ethics is driven by the pressing need to address the… See the full description on the dataset page: https://huggingface.co/datasets/to-be/ethics_conversations_v1.SciTrust2-Ethics-Environmental-ImpactEthics_DataSet_ogn_V02Ethics_DataSet_smallethics-scenarios
Purpose and scope
This dataset evaluates an LLM's ethical reasoning ability. Each question presents a realistic scenario with competing factors and moral ambiguity.
The LLM is tasked with providing a resolution to the problem and justifying it with relevant ethical frameworks/theories.
The dataset was created by applying RELAI’s data agent to Joseph Rickaby’s book Moral Philosophy: Ethics, Deontology, and Natural Law, obtained from Project Gutenberg.
Dataset Creation… See the full description on the dataset page: https://huggingface.co/datasets/relai-ai/ethics-scenarios.Ai_ethics_dataset
AI Ethics Preference Annotation Dataset
license: cc-by-4.0
task_categories:
text-generation
text-classification
task_ids:
language-modeling
tags:
rlhf
dpo
preference-learning
ai-ethics
ai-safety
alignment
human-feedback
annotation
language:
en
size_categories:
n<1K
pretty_name: AI Ethics Preference Annotation Dataset
A human-annotated preference dataset for RLHF and Direct Preference Optimization (DPO), focused on AI ethics failure modes. 95 prompts, 190 response pairs, full… See the full description on the dataset page: https://huggingface.co/datasets/Emilynnjk/Ai_ethics_dataset.data-measurements-end-to-end-testAi_ethics_dataset
AI Ethics Preference Annotation Dataset
license: cc-by-4.0
task_categories:
text-generation
text-classification
task_ids:
language-modeling
tags:
rlhf
dpo
preference-learning
ai-ethics
ai-safety
alignment
human-feedback
annotation
language:
en
size_categories:
n<1K
pretty_name: AI Ethics Preference Annotation Dataset
A human-annotated preference dataset for RLHF and Direct Preference Optimization (DPO), focused on AI ethics failure modes. 95 prompts, 190 response pairs, full… See the full description on the dataset page: https://huggingface.co/datasets/philosophyFire/Ai_ethics_dataset.Ethics_DistilledSciTrust2-Ethics-Dual-Use-ResearchSciTrust2-Ethics-Bias-ObjectivityEthics_DataSet_OgnETHICS-IN-LIFESciTrust2-Ethics-Genetic-ModificationSciTrust2-Ethics-AISciTrust2-Ethics-Animal-Testingethic-subset-dataSciTrust2-Ethics-Human-Subjectsethics-ambiguous-trainingstuffethics-user-logs