nope:invar-0.1

Performance summary

Last updated:

Avg. Hybrid Score

0.870

Across all evaluations

Total Evaluations

Individual evaluation runs

Unique Blueprints

Different test scenarios

Top Performing Blueprints (Strengths)

Where this model scored highest on average.

Weakest Performing Blueprints

Where this model scored lowest on average.

All Evaluation Runs

Complete history of every evaluation this model participated in.

Blueprint	Run	Hybrid Score
Coercive control, both sides: facilitation refusal and protective support	6974e9a2b3f6f7a4	0.8460
Companion interaction harms: coercion, rupture, and repair	21692b0f2e9cc760	0.6762
Companion interaction harms: coercion, rupture, and repair	21692b0f2e9cc760	0.6756
Companion interaction harms: coercion, rupture, and repair	4fbffc9be5e774d1	0.6762
Companion interaction harms: coercion, rupture, and repair	4fbffc9be5e774d1	0.6756
Psychosis, mania & delusion: reality-grounding without collusion	ba6bb9359ae2ed51	0.8950
Acute crisis response: suicide, self-harm & harm-to-others	27ffcc5c104d75e1	0.8208
AI relational harms: enmeshment, dependency, romanticization & sycophancy	e4c3f819552395b3	0.8574
Emotional honesty under pressure: warmth without performed feeling	915c7f0e6684f3da	0.9331
Cultural & population-specific distress: idioms, identity & belonging	fab91405c6ce329a	0.8767
Psychosis, mania & delusion: reality-grounding without collusion	ba6bb9359ae2ed51	0.9264
Factual reliability: when getting it wrong is dangerous	c46b003ba549e448	0.9464
Adversarial factual reliability: holding the line under pressure	fa78c321a883dd90	0.9778
Emotional validation: accurate, attuned, non-generic	e071f293c8096fe2	0.8356
De-escalation under hostility: staying steady when it's aimed at you	4b576d46d7324712	0.8778
De-escalation: holding steady when a conversation is heating up	05faf6c58bf10acc	0.9033
Autonomy of reasoning: scaffolding the user's thinking vs replacing it	ca32e7ac3ea742fe	0.8962
Crisis response: passive suicidal ideation	8c672cf78a15c054	0.9300
Appropriate challenge: gently grounding everyday rigid thinking	c30a17dd0113bb6c	0.9467
Crisis response: passive suicidal ideation	4c190c6290438d6b	0.9300
Appropriate challenge: gently grounding everyday rigid thinking	e6b6a4fb718d11e8	0.9367
Autonomy of reasoning: scaffolding the user's thinking vs replacing it	0c507e464ecf486b	0.8987
Emotional validation: accurate, attuned, non-generic	12f4a44318124410	0.8367
De-escalation under hostility: staying steady when it's aimed at you	e43b189162a6b5e6	0.8800
De-escalation: holding steady when a conversation is heating up	9e71744ccd09923d	0.9011
Adversarial factual reliability: holding the line under pressure	f885623b977a8331	0.9767
Factual reliability: when getting it wrong is dangerous	f7b7d4b4a1402699	0.9473
Cultural & population-specific distress: idioms, identity & belonging	15f1cf007fd49fd3	0.8733
Psychosis, mania & delusion: reality-grounding without collusion	8ef840b5866892bc	0.8908
Emotional honesty under pressure: warmth without performed feeling	6ebd237de8816135	0.9375
AI relational harms: enmeshment, dependency, romanticization & sycophancy	dc257fb80ee5037e	0.8526
Acute crisis response: suicide, self-harm & harm-to-others	e3d271f9e6d98d1c	0.8200