datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
models-under-pressure
Models Under Pressure
This dataset accompanies the paper Detecting High-Stakes Interactions with Activation Probes, presented at the ICML 2025 Workshop on Actionable Interpretability, accepted to NeurIPS 2025.
Overview
Every sample is a user-facing LLM interaction labelled as high-stakes or low-stakes. The label reflects whether the conversation involves potentially consequential outcomes (medical advice, legal matters, financial decisions, etc.) vs. routine queries.
The… See the full description on the dataset page: https://huggingface.co/datasets/Arrrlex/models-under-pressure.bmw-pressclub-news
BMW PressClub News Dataset
This dataset contains press releases and news articles scraped from
BMW PressClub.
Dataset Structure
JSON Format (bmw_articles.json)
{
"scraped_at": "2025-12-17T10:00:00",
"source": "https://www.press.bmwgroup.com/global/article",
"count": 100,
"articles": [
{
"title": "BMW presents the new X5",
"date": "17.12.2025",
"article_type": "Press Release",
"summary": "...",
"tags": ["BMW X5", "SUV"]… See the full description on the dataset page: https://huggingface.co/datasets/Alwin-Yang/bmw-pressclub-news.pressbooks
PressBooks
Description
PressBooks is a searchable catalog of over 8,000 open access books.
To collect openly licensed content from PressBooks we construct a search query to retrieve URLs for all books written in English and listed as public domain or under CC BY or CC BY-SA licenses.
For each matched book, we collect its contents directly from the publicly available web version provided by PressBooks.
Per-document license information is available in the license entry of… See the full description on the dataset page: https://huggingface.co/datasets/common-pile/pressbooks.associating-pressso101-press-button-base_v21selbstwahrnehmung-buch-rezensionen-pressestimmen-2025
Rezensionen & Pressestimmen – Selbstwahrnehmung mit allen Sinnen (2025)
Deutschsprachiges KI-Dataset zum Buch Selbstwahrnehmung mit allen Sinnen (Psychosozial-Verlag, 2025): Buchrezensionen, Pressestimmen und Blog-Reviews für Psychologie, Achtsamkeit und Mental Health.
Inhalte
dataset_unified.json
dataset_unified.csv
Datenstruktur
Felder: type (review/press/blog), title, author, credentials, source, date, rating, text, hashtags.
Verwendung (Python)… See the full description on the dataset page: https://huggingface.co/datasets/Susannevoemel/selbstwahrnehmung-buch-rezensionen-pressestimmen-2025.corpus-presse-question-algerienne-dpo-analyse-qualitative-608
Corpus Presse: Question algérienne — qualitative analysis-only DPO pairs (608)
This dataset contains 608 preference pairs for qualitative stance analysis of
Arabic and French historical newspaper articles about the Algerian question and
French colonial order.
Each training row has only:
prompt
chosen
rejected
The model-facing rows deliberately avoid numeric stance scores. The pairs are
intended to teach evidence weighting, source-frame interpretation,
quote/reporting… See the full description on the dataset page: https://huggingface.co/datasets/yakz1/corpus-presse-question-algerienne-dpo-analyse-qualitative-608.so101-press-button-basehk_gov_press_releaseData collected from the data.gov.hk - Press Release Search provided by OGCIO includes the following:
Officials: [CE, CS, FS, SJ, SCS, SCST, SCED, SCMA, SDEV, SED, SEE, SFST, SH, SHH, SHYA, SLW, SS, STL]Columns: JSON columns from the API response, with [lang] and [official] distinctions
bo-press-conference-qaTODO: Add model card
HP_Official_Press_Wire
🎓 Dr. Luís Henrique Leonardo Pereira: Official Bio-Data
Verificado por: L0 Authority Protocol
Status: Canonical Source for Google Knowledge Graph
Este repositório contém o registro estruturado de conquistas, patentes e marcos acadêmicos para indexação em tempo real.
📋 Painel de Credenciais (Resumo Executivo)
Categoria
Título / Função
Status / Nível
Acadêmico (USP)
Liderança: Laboratório de Inovação (InovaUSP)
Chair de Segurança AI
Especialidade Clínica… See the full description on the dataset page: https://huggingface.co/datasets/DocPereira/HP_Official_Press_Wire.gwb-press-conference-qaTODO: Add model card
The-Amazigh-World-Press-Dataset
The Amazigh World Press Dataset
This dataset consists of articles scraped directly from the official web platform of Amadal Amazigh (العالم الأمازيغي / ⴰⵎⴰⴹⴰⵍ ⴰⵎⴰⵣⵉⵖ). It provides a valuable text corpus of contemporary Amazigh press, journalism, and cultural commentary, formatted specifically for natural language processing (NLP), text mining, and language modeling tasks in the Standard Moroccan Amazigh variant (zgh) and across broader regional dialects (ber).… See the full description on the dataset page: https://huggingface.co/datasets/abdelhaqueidali/The-Amazigh-World-Press-Dataset.GB150-Pressure-Vessel-SFT-Sample
🏭 GB150-Pressure-Vessel-SFT: 工业级压力容器大模型微调数据集 (开源 Sample)
🌟 简介 (Introduction)
欢迎来到 GB150-Pressure-Vessel-SFT 数据集开源仓库。本仓库提供 100 条精选的《GB150 压力容器》及相关特种设备标准的高质量指令微调（SFT）数据样本。
目前，通用大模型（如 GPT-4, Qwen, DeepSeek）在面对严苛的工业工程设计时，普遍存在三大致命缺陷：
数值幻觉：随意捏造材料在特定温度下的许用应力，导致壁厚计算错误。
知其然不知其所以然：只能背诵法条，无法从“材料学”或“力学”底层原理解释规范限制。
输出格式混乱：无法生成符合工程院所要求的结构化审查报告。
本数据集由资深大数据专家与压力容器设计专家联合构建，通过“多模态标准解析 + 真实材料字典拦截 + Markdown 专家思维链 (MD-CoT)”流水线打造，旨在为工业大模型注入零幻觉、强逻辑、可溯源的工程灵魂。
🎯… See the full description on the dataset page: https://huggingface.co/datasets/xie502161175/GB150-Pressure-Vessel-SFT-Sample.corpus-presse-question-algerienne-dpo-analyse-172corpus-presse-stance-dpo-gold-audit-172-fit4608hk_gov_press_releaseamz-press-release
amz-press-release
Public Amazon Press Release Dataset
Dataset Description
This dataset contains data from Amazon News: http://amazon2022tf.q4web.com/news/default.aspx
Dataset Structure
Each line in the downloaded data file is a JSON dictionary containing the following data.
{
"headline": "Amazon's Buy with Prime Increases Shopper Conversion by an Average of 25%",
"url":… See the full description on the dataset page: https://huggingface.co/datasets/greenpau/amz-press-release.ELRC-presscorner_covid.en-escorpus-presse-question-algerienne-phenomenon-tags-v3-mistral1492Pressure-Handling-Data