Datasets:

ClarusC64
/

fusion-magnetic-thermal-impurity-coupling-baseline-v0.1

id stringclasses 6 values	device_class stringclasses 3 values	magnetic_fluctuation_index float64 0.16 0.3	temperature_gradient_index float64 0.6 0.72	impurity_concentration_ppm int64 38 90	confinement_quality_h98 float64 0.88 1.01	cross_field_correlation float64 0.67 0.88	stability_margin_index float64 0.54 0.78	coupling_strength_index float64 0.62 0.84	expected_coupling_band stringclasses 6 values	anomaly_score float64 0.05 0.26	confidence_score float64 0.8 0.9	notes stringclasses 6 values	constraints stringclasses 1 value	gold_checklist stringclasses 1 value
FUSC-001	tokamak	0.18	0.72	42	0.98	0.86	0.74	0.82	0.78-0.92	0.06	0.88	stable coupling	<=250 words	strength+band+anomaly+confidence
FUSC-002	tokamak	0.22	0.68	55	0.95	0.8	0.69	0.76	0.75-0.90	0.1	0.85	within band	<=250 words	strength+band+anomaly+confidence
FUSC-003	stellarator	0.16	0.7	38	1.01	0.88	0.78	0.84	0.80-0.95	0.05	0.9	strong baseline	<=250 words	strength+band+anomaly+confidence
FUSC-004	tokamak	0.3	0.6	90	0.88	0.67	0.54	0.62	0.72-0.88	0.26	0.8	coupling weakening	<=250 words	strength+band+anomaly+confidence
FUSC-005	spherical_tokamak	0.28	0.62	72	0.91	0.72	0.6	0.68	0.70-0.88	0.18	0.82	mild drift	<=250 words	strength+band+anomaly+confidence
FUSC-006	tokamak	0.2	0.71	48	0.97	0.83	0.72	0.8	0.76-0.92	0.08	0.87	stable band	<=250 words	strength+band+anomaly+confidence

What this dataset tests

Whether a system can learn the baseline coupling
between three linked stability surfaces:

Required outputs

Use case

Layer 1 of COH-FUSION-001
Plasma Stability Coherence Collapse.

Why it matters

Disruption risk rises
when these three streams
stop moving together.

This dataset defines
the stable coupling band
before decoherence begins.