HuggingFaceFW/fineweb-2
FineWeb-2 is a dataset of over 15 trillion tokens of cleaned and deduplicated English web data from CommonCrawl. This is the second iteration of the popular 🍷 FineWeb dataset, bringing high quality pretraining data to over 1000 🗣️ languages.The 🥂 FineWeb2 dataset is fully reproducible, available under the permissive ODC-By 1.0 license and extensively validated through hundreds of ablation experiments.In particular, on the set of 9 diverse languages we used to guide our processing decisions, 🥂 FineWeb2 outperforms other popular pretraining datasets covering multiple languages (such as CC-100, mC4, CulturaX or HPLT, while being substantially larger) and, in some cases, even performs better than some datasets specifically curated for a single one of these languages, in our diverse set of carefully selected evaluation tasks: FineTasks.
Detaylı Giriş
FineWeb-2 is a large-scale dataset designed to provide high-quality web data for training large language models. This is the second iteration of the popular 🍷 FineWeb dataset, bringing high quality pretraining data to over 1000 🗣️ languages.The 🥂 FineWeb2 dataset is fully reproducible, available under the permissive ODC-By 1.0 license and extensively validated through hundreds of ablation experiments.In particular, on the set of 9 diverse languages we used to guide our processing decisions, 🥂 FineWeb2 outperforms other popular pretraining datasets covering multiple languages (such as CC-100, mC4, CulturaX or HPLT, while being substantially larger) and, in some cases, even performs better than some datasets specifically curated for a single one of these languages, in our diverse set of carefully selected evaluation tasks: FineTasks.
Daha fazlaVeri seti
SAMHDA: Substance Abuse and Mental Health Data Archive
The Substance Abuse and Mental Health Data Archive (SAMHDA) provides a comprehensive collection of data sets related to mental health and substance use. It includes ongoing studies, population surveys, treatment facility surveys, and client-level data, offering valuable insights for researchers and policymakers.
HeartLink - Empathetic Psychological Model
HeartLink is an empathetic psychological model that uses a large language model fine-tuned on a large empathetic Q&A dataset. It can perceive users' emotions and experiences during conversations and provide empathetic responses using rich psychological knowledge, aiming to understand, comfort, and support users. The responses include emoji expressions to bridge the gap with users, offering psychological support and help during consultations.
Data Science for COVID-19 (DS4C)
The DS4C dataset is a structured collection of COVID-19 data from South Korea, based on reports from the Korea Centers for Disease Control & Prevention (KCDC) and local governments. It includes information on infections, patient routes, and various analyses. The dataset has been used for multiple research and visualization projects.