datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
scramble-control-panels
Scramble-control panels for cofolding confidence metrics
Per-fold confidence scores for peptide–protein complexes, folded under Boltz-1,
Boltz-2, Chai-1 and a few-step-distilled model, with each cognate peptide scored
against permutations of itself as well as against unrelated decoys.
2,456 folds across 16 inference arms and 75 receptors.
A permutation — a scramble — preserves amino-acid composition and length
exactly and destroys only sequence order. Decoy comparisons cannot… See the full description on the dataset page: https://huggingface.co/datasets/AkikJana/scramble-control-panels.ethical-ai-control-panel-synthetic-risk
Synthetic Agent Risk Dataset v1
This dataset contains a fully synthetic collection of short English prompts that describe AI agents or automation workflows.Each prompt is labeled with one of three risk categories:
low_risk
medium_risk
high_risk
The dataset was created for an educational project (Ethical AI Control Panel) in a university course on responsible machine learning.It is intended to support proof‑of‑concept risk classifiers, not to represent real‑world deployment data.… See the full description on the dataset page: https://huggingface.co/datasets/EricCRX/ethical-ai-control-panel-synthetic-risk.