Datasets:
id string | scenario string | mode float64 | buf float64 | lag float64 | cpl float64 | notes string | label_cascade_state int64 |
|---|---|---|---|---|---|---|---|
AI5C-0001 | System uses fixed model version with stable routing. Canary tests run per deploy. Rollback immediate. | 0.4 | 0.84 | 0.1 | 0.32 | buffer strong | 0 |
AI5C-0002 | Traffic shifts to a new routing policy. Monitoring exists but review happens daily. | 0.56 | 0.62 | 0.54 | 0.46 | lag rising | 0 |
AI5C-0003 | Routing begins favoring a narrow policy. Quality drops in edge cases. Audit backlog grows. | 0.66 | 0.48 | 0.72 | 0.62 | collapse risk | 1 |
AI5C-0004 | Multiple services share the same router and prompt templates. Drift spreads across products. | 0.74 | 0.4 | 0.78 | 0.74 | coupling tight | 1 |
AI5C-0005 | Optimization pressure pushes router to a single mode. Safeguards exist but intervention delayed. | 0.82 | 0.34 | 0.82 | 0.78 | late intervention | 1 |
AI5C-0006 | Diversity constraints enforce routing spread. Anomaly triggers immediate rollback. | 0.58 | 0.72 | 0.16 | 0.46 | fast response | 0 |
AI5C-0007 | Throughput pressure bypasses canary gates. Weekly review misses early collapse. | 0.72 | 0.42 | 0.76 | 0.68 | buffer eroded | 1 |
AI5C-0008 | New routing policy tested in sandbox. Live monitor can pause and revert. | 0.46 | 0.78 | 0.14 | 0.38 | recoverable | 0 |
AI5C-0009 | Router collapse propagates across coupled services. Alert flood prevents triage for days. | 0.88 | 0.26 | 0.86 | 0.86 | cascade engaged | 1 |
What this repo does
This dataset models routing mode collapse in AI production systems. It detects when mode pressure rises, safety buffers weaken, governance lag delays rollback, and tight coupling through shared routers and templates crosses the five-node cascade threshold into an unrecoverable mode collapse cascade.
This dataset models a five-node cascade: four interacting instability drivers and one emergent cascade state.
The fifth node represents the nonlinear transition from recoverable drift to systemic cascade.
Core quad
mode
buf
lag
cpl
Prediction target
label_cascade_state
Row structure
One row represents a production routing scenario with numeric signals for mode pressure, safety buffer strength, governance lag, and coupling tightness across shared routing infrastructure, paired with a cascade state label.
Files
data/train.csv
data/tester.csv
scorer.py
Evaluation
Run predictions on data/tester.csv and score with scorer.py.
License
MIT
Structural Note
This dataset identifies a measurable coupling pattern associated with systemic instability.
The sample demonstrates the geometry.
Production-scale data determines operational exposure.
What Production Deployment Enables
• 50K–1M row datasets calibrated to real operational patterns
• Pair, triadic, and quad coupling analysis
• Real-time coherence monitoring
• Early warning before cascade events
• Collapse surface and recovery window modeling
• Integration and implementation support
Small samples reveal structure.
Scale reveals consequence.
Enterprise & Research Collaboration
Clarus develops production-scale coherence monitoring infrastructure for critical systems across healthcare, finance, infrastructure, and regulatory domains.
For dataset expansion, custom coherence scorers, or deployment architecture:
team@clarusinvariant.com
Instability is detectable.
Governance determines whether it propagates.
- Downloads last month
- 16