Views
No views yet
PURE: Pretraining Universal Reasoning Engine 🧠
Dataset Name: PURECurator: Xerv-AITarget Architecture: Optimized for training highly efficient reasoning models (500M - 3B parameters) entirely from scratch.Language: English (Predominantly), with multi-lingual STEM subsets.License: Mixed (Derived from respective source datasets; strictly for non-commercial research purposes where applicable).
📖 Dataset Overview
The PURE (Pretraining Universal… See the full description on the dataset page: https://huggingface.co/datasets/Xerv-AI/PURE.