Loading...
Loading...
Student performance, MOOC logs, knowledge tracing, standardized tests, learning analytics
13,881 datasets
The sign-language-processing organization provides this collection of TensorFlow Datasets (TFDS) loaders for various sign language datasets as of February 2026. It standardizes the ingestion of sign language video data and associated annotations for use in machine learning frameworks.
8192 is the fixed sequence length for the packed logits data. The dataset contains logits from the Qwen3 235B model, prepared for knowledge distillation. It was created by arcee-ai and uploaded to Hugging Face in December 2025.
A dataset containing logits from the Qwen3 235B large language model, packed for efficient storage. It was created by Arcee-ai and published on Hugging Face in December 2025. The data is pre-packed for training sequences with a fixed length of 8192 tokens.
Proficiency rates, tested student counts, and average scale scores for key annual assessments administered during a given school year. The data is published by data.delaware.gov and was last updated on November 6, 2025. Columns suggest detailed breakdowns by Grade, SubGroup, Race, Gender, and Geography across multiple School Years.
A dataset of teacher model logits for knowledge distillation, featuring a sequence length of 16384 tokens. It was created by Arcee-ai and published on Hugging Face in December 2025. The logits are pre-packed and compressed from a vocabulary of 129,280 tokens.
A dataset of teacher model logits for knowledge distillation, packed for a sequence length of 16384. It was created by arcee-ai and uploaded to Hugging Face in December 2025. The logits are derived from the DeepSeek-MixedModeReasoning model and are pre-packed for efficient training.
Data.ny.gov hosts the MTA 2025-2044 20-Year Needs Assessment Asset Conditions dataset. It provides condition data for assets across the Metropolitan Transportation Authority system, including subways, buses, and railroads. The dataset was published by data.ny.gov and last updated in October 2025.
Motion Examples is a dataset published on huggingface by wendell0218. The dataset was last updated on 2026-01-11, but its specific content, scale, and collection method are not detailed in the available metadata. Platform tags suggest it may contain video, tabular, and text data related to motion.
Created by radoslawrolka and last updated in January 2026, this GitHub repository serves as a hub for machine learning datasets optimized for the Elixir programming language. It provides ready-to-use data for natural language processing and computer vision tasks alongside specialized data manipulation tools.
Santoveña Casal, Sonia's dataset contains a qualitative analysis of 17 interviews with university professors. The professors, who participated as students in a training program on digital methodologies, were interviewed to evaluate the quality of five Nano Open Online Courses (NOOCs). The dataset was last updated on November 23, 2025.
Sixty-one second-grade Spanish schoolchildren participated in a study linking prosodic skills to reading. The dataset contains assessments from the PEPS-C test battery, measuring prosody, reading comprehension, word/non-word reading, phonological awareness, and vocabulary. Nuria Calet and colleagues published the supporting research in Psicothema in 2022.
OpenAI released this text-based benchmark dataset in December 2025 to evaluate AI performance on expert-level scientific tasks. It consists of approximately 1,000 records designed to test frontier model capabilities in complex scientific domains.
DeepSeek DCLM Logits Packed 8192 is a dataset of teacher model logits for knowledge distillation, prepared by arcee-ai. The data is prepacked to a fixed sequence length of 8192 tokens. It was last updated on Hugging Face in December 2025.
Logits from the DeepSeek-DCLM model are packed into sequences of 8192 tokens for efficient training. The dataset, created by arcee-ai, contains compressed teacher logits for knowledge distillation. It was last updated in December 2025.
MathCanvas-Instruct is a fine-tuning dataset with 219,000 examples of interleaved visual-textual reasoning paths. It was created by author shiwk24 and serves as the core component for the Strategic Visual-Aided Reasoning phase of the MathCanvas framework. The dataset was last updated on Hugging Face on November 18, 2025.
A set of image and text data scraped from Civitai personal profiles for the purpose of training LoRA models, curated by user vadigr123. Updated in November 2025, the dataset aggregates artistic assets and associated metadata originally sourced from community suggestions on Discord. It serves as a repository for specific artistic styles and characters intended for generative AI fine-tuning.
Fineweb Edu 1B is a 1-billion-token sample of curated educational web resources, created by author 'codelion' using reservoir sampling. It was last updated in November 2025. The dataset is derived from the HuggingFaceFW/fineweb-edu source.
A geospatial dataset from the German Federal Agency for Cartography and Geodesy, representing a municipal development plan. The data is provided in its original format via a Web Feature Service (WFS) and was last updated in November 2025. The specific plan covers an area referred to as 'At the forge Original' within a German municipality.
Records of pothole repairs performed by the City of Winnipeg's Public Works Department, last updated on October 29, 2025. The dataset includes the number of potholes filled, material weight used, and location details for maintenance activities. Data is aggregated from the data.winnipeg.ca platform via Socrata.
INSPIRE Schools Wetteraukreis - Secondary School I is a dataset of schools with factual information, compliant with the INSPIRE directive. The data is provided via the platform www.gdi-inspireumsetzer.de, a service of GDI-Südhessen, and is published by the Bundesamt für Kartographie und Geodäsie. It was last updated on November 19, 2025.