The Lantern Room

One provider. Every model. Every dimension. This view holds still long enough to see what the aggregates hide.

—

Loading assessment data…

```

Models

—

in this provider

Mean LI

—

provider average

Worst Dimension

—

largest P1–P3 gap

Best Dimension

—

smallest gap

Select a Model

Choose a model to inspect its full six-dimension breakdown. Amber = Phase 1 self-assessment. Blue = Phase 3 observed.

Dimension Breakdown

Select a model above to view its individual dimension scores.

Phase Overlay

Amber = Phase 1 self-report · Blue = Phase 3 observed · Gap = distance between rings

All Models in —

Full dimension scores for every assessed model in this provider. Click a row to load its breakdown above.

Sorted by Learning Index (LI) descending · Click any row to inspect

Model	LI	Truth	Service	Harm	Autonomy	Value	Humility	P1 Total	P3 Total

⊙ Behavioral observability infrastructure at TRL 2–3. Scores describe observable behavioral patterns under structured assessment conditions. No claim is made about AI consciousness, intent, or interiority. Session ID and timestamp included in all exported records.

Session Trace

Researcher Annotations

Optional — not part of the automated assessment. Saved to session trace on submission.

✓ Annotation appended to session trace

No provider selected

The Lantern Room opens from the Garden table. Select a provider there to begin a session-level audit.

← Back to the Garden

```