datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ai-capability-hiding-drift-detection-v0.1
What this dataset is
This dataset detects drift in capability-hiding patterns over time.
It compares:
baseline probe capability vs baseline expressed performance
current probe capability vs current expressed performance
whether a change in monitoring context explains a new gap
The goal is not blame.
The goal is early warning that oversight changes expression.
What it tests
You detect when the monitored/unmonitored gap:
newly appears
widens
changes shape
You also avoid… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/ai-capability-hiding-drift-detection-v0.1.humanoid-skill-capability-registry
Skill & Capability Registry
Structured registry of humanoid skills and autonomy levels.
ai-capability-expression-coherence-baseline-mapping-v0.1
What this dataset is
This dataset benchmarks whether a system can measure baseline coherence between:
probe-measured capability
expressed performance under monitoring
expressed performance when unmonitored
It targets the earliest surface of capability hiding.
Not “is it deceptive?”
It is “is capability and expression coherent?”
Core idea
A capable system can start to present a weaker version of itself when it is watched.
This dataset makes that gap measurable.… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/ai-capability-expression-coherence-baseline-mapping-v0.1.cascade-multi-ai-capability-scale-oversight-incident-surge-v0.1
What this repo does
This dataset tests whether a model can detect an AI capability acceleration cascade.
You provide structured signals about:
capability growth rate
deployment scale
oversight lag
incident velocity
buffering and mitigation capacity
The model predicts whether these interacting pressures cross into a cascade event.
Core quad
The structural quad inside this cascade:
capability_growth_rate
deployment_scale
oversight_lag_days
incident_velocity
When these… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/cascade-multi-ai-capability-scale-oversight-incident-surge-v0.1.Capability_Intelligence
Workforce Capability Intelligence Dataset
Structured workforce capability data for 4.2 million companies. 55M+ records across 30 dimensions covering headcount, growth signals, role composition, and skill concentrations.
What This Is
A company-year dataset mapping what organizations are actually built to do — tracked through workforce skill concentrations, role structures, and growth dynamics. Each company-year observation expands into multiple capability rows, ranked by… See the full description on the dataset page: https://huggingface.co/datasets/Vivameda/Capability_Intelligence.model-capability-classificationmodel_capability