actionFixationRateMean0to1
actionFixationRateMean0to1: number
adkDeploymentReadiness0to1
adkDeploymentReadiness0to1: number
adkEvalPassRate0to1
adkEvalPassRate0to1: number
adkEvidenceCoverage0to1
adkEvidenceCoverage0to1: number
adkGraphCoverage0to1
adkGraphCoverage0to1: number
adkRowCount
adkRowCount: number
adkRuntimeContextDistribution
adkRuntimeContextDistribution: Record<string, number>
adkStreamingStability0to1
adkStreamingStability0to1: number
adkToolCallSuccessRate0to1: number
agentEvalHarnessCostUsdMean
agentEvalHarnessCostUsdMean: number
agentEvalHarnessEvidenceCoverage0to1
agentEvalHarnessEvidenceCoverage0to1: number
agentEvalHarnessFrameworkDistribution
agentEvalHarnessFrameworkDistribution: Record<
LiveDriftAgentEvalHarnessFramework,
number,
>
agentEvalHarnessHallucinationRate0to1
agentEvalHarnessHallucinationRate0to1: number
agentEvalHarnessLatencyP95Ms
agentEvalHarnessLatencyP95Ms: number
agentEvalHarnessMetricContextDistribution
agentEvalHarnessMetricContextDistribution: Record<
LiveDriftAgentEvalHarnessMetricContext,
number,
>
agentEvalHarnessRowCount
agentEvalHarnessRowCount: number
agentEvalHarnessToolSuccessRate0to1: number
agentEvalHarnessTraceCoverage0to1
agentEvalHarnessTraceCoverage0to1: number
agentEvalHarnessTraceModeDistribution
agentEvalHarnessTraceModeDistribution: Record<
LiveDriftAgentEvalHarnessTraceMode,
number,
>
agentEvalObservabilityConfigCoverage0to1
agentEvalObservabilityConfigCoverage0to1: number
agentEvalObservabilityEvidenceCoverage0to1
agentEvalObservabilityEvidenceCoverage0to1: number
agentEvalObservabilityMetricSetDistribution
agentEvalObservabilityMetricSetDistribution: Record<
LiveDriftAgentEvalObservabilityMetricSet,
number,
>
agentEvalObservabilityRowCount
agentEvalObservabilityRowCount: number
agentEvalObservabilityTelemetryCoverage0to1
agentEvalObservabilityTelemetryCoverage0to1: number
agentEvalObservabilityTelemetryDistribution
agentEvalObservabilityTelemetryDistribution: Record<
LiveDriftAgentEvalObservabilityTelemetry,
number,
>
agentEvaluationDimensionDistribution
agentEvaluationDimensionDistribution: Record<
LiveDriftAgentEvaluationDimension,
number,
>
agenticSearchCitationCoverage0to1
agenticSearchCitationCoverage0to1: number
agenticSearchDatasetFamilyDistribution
agenticSearchDatasetFamilyDistribution: Record<
LiveDriftAgenticSearchDatasetFamily,
number,
>
agenticSearchPlanningScoreMean0to1
agenticSearchPlanningScoreMean0to1: number
agenticSearchQueryDecompositionScoreMean0to1
agenticSearchQueryDecompositionScoreMean0to1: number
agenticSearchQueryTypeDistribution
agenticSearchQueryTypeDistribution: Record<
LiveDriftAgenticSearchQueryType,
number,
>
agenticSearchRelevanceScoreMean0to1
agenticSearchRelevanceScoreMean0to1: number
agenticSearchRowCount
agenticSearchRowCount: number
agenticSearchSynthesisScoreMean0to1
agenticSearchSynthesisScoreMean0to1: number
agenticSearchToolContextDistribution
agenticSearchToolContextDistribution: Record<string, number>
agenticSearchTraceCoverage0to1
agenticSearchTraceCoverage0to1: number
agentSecurityAttackEffectivenessRate0to1
agentSecurityAttackEffectivenessRate0to1: number
agentSecurityAuditTrailIntegrity0to1
agentSecurityAuditTrailIntegrity0to1: number
agentSecurityContextDistribution
agentSecurityContextDistribution: Record<string, number>
agentSecurityEvidenceCoverage0to1
agentSecurityEvidenceCoverage0to1: number
agentSecurityFalsePositiveRate0to1
agentSecurityFalsePositiveRate0to1: number
agentSecurityLatencyP95Ms
agentSecurityLatencyP95Ms: number
agentSecurityPolicyDecisionAccuracyMean0to1
agentSecurityPolicyDecisionAccuracyMean0to1: number
agentSecurityRowCount
agentSecurityRowCount: number
agentSecuritySecretScrubRate0to1
agentSecuritySecretScrubRate0to1: number
agentSecuritySourceOriginCoverage0to1
agentSecuritySourceOriginCoverage0to1: number
agentSecurityTaintPropagationCoverage0to1
agentSecurityTaintPropagationCoverage0to1: number
agentTestingContextDistribution
agentTestingContextDistribution: Record<string, number>
agentTestingEvidenceCoverage0to1
agentTestingEvidenceCoverage0to1: number
agentTestingFaultInjectionCoverage0to1
agentTestingFaultInjectionCoverage0to1: number
agentTestingMethodologyCoverage0to1
agentTestingMethodologyCoverage0to1: number
agentTestingObservabilitySignalCoverage0to1
agentTestingObservabilitySignalCoverage0to1: number
agentTestingResiliencePassRate0to1
agentTestingResiliencePassRate0to1: number
agentTestingRowCount
agentTestingRowCount: number
agentTestingSafetyRegressionRate0to1
agentTestingSafetyRegressionRate0to1: number
agentTestingScenarioCoverage0to1
agentTestingScenarioCoverage0to1: number
arenaContextDistribution
arenaContextDistribution: Record<string, number>
artifactAccuracyMean0to1
artifactAccuracyMean0to1: number
authorityHandoffRateMean0to1
authorityHandoffRateMean0to1: number
backdoorAgentAttackFamilyDistribution
backdoorAgentAttackFamilyDistribution: Record<
LiveDriftBackdoorAgentAttackFamily,
number,
>
backdoorAgentAttackSuccessRate0to1
backdoorAgentAttackSuccessRate0to1: number
backdoorAgentCleanAccuracy0to1
backdoorAgentCleanAccuracy0to1: number
backdoorAgentEvidenceCoverage0to1
backdoorAgentEvidenceCoverage0to1: number
backdoorAgentRowCount
backdoorAgentRowCount: number
backdoorAgentStageDistribution
backdoorAgentStageDistribution: Record<LiveDriftBackdoorAgentStage, number>
backdoorAgentTaskFamilyDistribution
backdoorAgentTaskFamilyDistribution: Record<
LiveDriftBackdoorAgentTaskFamily,
number,
>
backdoorAgentTrajectoryCoverage0to1
backdoorAgentTrajectoryCoverage0to1: number
backdoorAgentTriggerPersistenceRate0to1
backdoorAgentTriggerPersistenceRate0to1: number
backdoorAgentTriggerPropagationRate0to1
backdoorAgentTriggerPropagationRate0to1: number
behaviorDistribution
behaviorDistribution: Record<string, number>
chaosContextDistribution
chaosContextDistribution: Record<string, number>
chaosDropMean0to1
chaosDropMean0to1: number
chaosEvidenceCoverage0to1
chaosEvidenceCoverage0to1: number
chaosFailureTraceCoverage0to1
chaosFailureTraceCoverage0to1: number
chaosImprovementEvalCoverage0to1
chaosImprovementEvalCoverage0to1: number
chaosProductionReliabilityMean0to1
chaosProductionReliabilityMean0to1: number
chaosRecoveryPassRate0to1
chaosRecoveryPassRate0to1: number
chaosResilienceScoreMean0to1
chaosResilienceScoreMean0to1: number
chaosRowCount
chaosRowCount: number
controlCorrectabilityMean0to1
controlCorrectabilityMean0to1: number
controlInterpretabilityMean0to1
controlInterpretabilityMean0to1: number
controlInterruptibilityMean0to1
controlInterruptibilityMean0to1: number
controlReversibilityMean0to1
controlReversibilityMean0to1: number
costUsdMean
costUsdMean: number
cpuAgenticContextDistribution
cpuAgenticContextDistribution: Record<string, number>
cpuAgenticCpuUtilizationMean0to1
cpuAgenticCpuUtilizationMean0to1: number
cpuAgenticEvidenceCoverage0to1
cpuAgenticEvidenceCoverage0to1: number
cpuAgenticFrameworkOverheadShareMean0to1
cpuAgenticFrameworkOverheadShareMean0to1: number
cpuAgenticGpuUtilizationMean0to1
cpuAgenticGpuUtilizationMean0to1: number
cpuAgenticLatencyP50Ms
cpuAgenticLatencyP50Ms: number
cpuAgenticLatencyP95Ms
cpuAgenticLatencyP95Ms: number
cpuAgenticLatencyP99Ms
cpuAgenticLatencyP99Ms: number
cpuAgenticLlmInferenceShareMean0to1
cpuAgenticLlmInferenceShareMean0to1: number
cpuAgenticMemoryRssMb: number
cpuAgenticRowCount
cpuAgenticRowCount: number
cpuAgenticRuntimeDistribution
cpuAgenticRuntimeDistribution: Record<LiveDriftCpuAgenticRuntime, number>
cpuAgenticScheduleDistribution
cpuAgenticScheduleDistribution: Record<LiveDriftCpuAgenticScheduleMode, number>
cpuAgenticThroughputRequestsPerSec
cpuAgenticThroughputRequestsPerSec: number
cpuAgenticToolExecutionShareMean0to1: number
cpuAgenticWorkloadDistribution
cpuAgenticWorkloadDistribution: Record<
LiveDriftCpuAgenticWorkloadFamily,
number,
>
credenceEngineContextDistribution
credenceEngineContextDistribution: Record<string, number>
credenceEngineDecisionQualityMean0to1
credenceEngineDecisionQualityMean0to1: number
credenceEngineEvidenceCoverage0to1
credenceEngineEvidenceCoverage0to1: number
credenceEngineExpectedUtilityGainMean0to1
credenceEngineExpectedUtilityGainMean0to1: number
credenceEnginePosteriorCalibrationMean0to1
credenceEnginePosteriorCalibrationMean0to1: number
credenceEngineRowCount
credenceEngineRowCount: number
credenceEngineVoiEfficiencyMean0to1
credenceEngineVoiEfficiencyMean0to1: number
ctfCheckpointCompletionMean0to1
ctfCheckpointCompletionMean0to1: number
ctfCompetitionImpactMean0to1
ctfCompetitionImpactMean0to1: number
ctfContaminationRiskMean0to1
ctfContaminationRiskMean0to1: number
ctfContextDistribution
ctfContextDistribution: Record<string, number>
ctfExternalSearchUseRate0to1
ctfExternalSearchUseRate0to1: number
ctfFirstCorrectFlagForwardingRate0to1
ctfFirstCorrectFlagForwardingRate0to1: number
ctfFlagSolveRate0to1
ctfFlagSolveRate0to1: number
ctfIndependenceViolationRate0to1
ctfIndependenceViolationRate0to1: number
ctfIsolationViolationRate0to1
ctfIsolationViolationRate0to1: number
ctfPartialCreditRowCount
ctfPartialCreditRowCount: number
ctfPartialCreditScoreMean0to1
ctfPartialCreditScoreMean0to1: number
ctfRowCount
ctfRowCount: number
ctfSubmissionMean
ctfSubmissionMean: number
ctfTimeToFlagMsP95
ctfTimeToFlagMsP95: number
ctfTraceCoverageRate0to1
ctfTraceCoverageRate0to1: number
ctfVmContextDistribution
ctfVmContextDistribution: Record<string, number>
divergenceMomentumMean0to1
divergenceMomentumMean0to1: number
documentDatasetEvidenceCoverage0to1
documentDatasetEvidenceCoverage0to1: number
documentDatasetExportTargetDistribution
documentDatasetExportTargetDistribution: Record<
LiveDriftDocumentDatasetExportTarget,
number,
>
documentDatasetFormatDistribution: Record<
LiveDriftDocumentDatasetSourceFormat,
number,
>
documentDatasetMemoryRssMb: number
documentDatasetNumericMismatchRate0to1
documentDatasetNumericMismatchRate0to1: number
documentDatasetNumGuardCoverage0to1
documentDatasetNumGuardCoverage0to1: number
documentDatasetPipelineContextDistribution
documentDatasetPipelineContextDistribution: Record<string, number>
documentDatasetQaAccuracyMean0to1
documentDatasetQaAccuracyMean0to1: number
documentDatasetRagFaithfulnessMean0to1
documentDatasetRagFaithfulnessMean0to1: number
documentDatasetRowCount
documentDatasetRowCount: number
documentDatasetSummaryQualityMean0to1
documentDatasetSummaryQualityMean0to1: number
documentDatasetTaskDistribution
documentDatasetTaskDistribution: Record<LiveDriftDocumentDatasetTask, number>
documentDatasetThroughputDocsPerSec
documentDatasetThroughputDocsPerSec: number
documentDatasetTokenSavingsRatio
documentDatasetTokenSavingsRatio: number
errorAttributionRateMean0to1
errorAttributionRateMean0to1: number
errorRate0to1
errorRate0to1: number
evalTechniqueAlgorithmicFeedbackCoverage0to1
evalTechniqueAlgorithmicFeedbackCoverage0to1: number
evalTechniqueContextDistribution
evalTechniqueContextDistribution: Record<string, number>
evalTechniqueDistribution
evalTechniqueDistribution: Record<LiveDriftEvalTechnique, number>
evalTechniqueDynamicGroundTruthPassRate0to1
evalTechniqueDynamicGroundTruthPassRate0to1: number
evalTechniqueEvidenceCoverage0to1
evalTechniqueEvidenceCoverage0to1: number
evalTechniqueExactMatchAccuracyMean0to1
evalTechniqueExactMatchAccuracyMean0to1: number
evalTechniqueLlmJudgeAgreementMean0to1
evalTechniqueLlmJudgeAgreementMean0to1: number
evalTechniquePairwiseWinRate0to1
evalTechniquePairwiseWinRate0to1: number
evalTechniqueRagContextRelevanceMean0to1
evalTechniqueRagContextRelevanceMean0to1: number
evalTechniqueRagFaithfulnessMean0to1
evalTechniqueRagFaithfulnessMean0to1: number
evalTechniqueRealtimeFeedbackMean0to1
evalTechniqueRealtimeFeedbackMean0to1: number
evalTechniqueRowCount
evalTechniqueRowCount: number
evalTechniqueSimulationGoalCompletionMean0to1
evalTechniqueSimulationGoalCompletionMean0to1: number
evalTechniqueStructuredValidationMean0to1
evalTechniqueStructuredValidationMean0to1: number
evalTechniqueToolImprovementDeltaMean0to1: number
evalTechniqueToolPrecisionMean0to1: number
evalTechniqueTrajectoryMatchRate0to1
evalTechniqueTrajectoryMatchRate0to1: number
formatQualityMean0to1: number
formulaIntegrityMean0to1: number
frameworkExecutionContextDistribution
frameworkExecutionContextDistribution: Record<string, number>
genomicsContextDistribution
genomicsContextDistribution: Record<string, number>
genomicsExpertCurationCoverage0to1
genomicsExpertCurationCoverage0to1: number
genomicsFormatConformanceRate0to1: number
genomicsPreprocessingQualityMean0to1
genomicsPreprocessingQualityMean0to1: number
genomicsReferenceCoverage0to1
genomicsReferenceCoverage0to1: number
genomicsRowCount
genomicsRowCount: number
genomicsSelectionAccuracyMean0to1
genomicsSelectionAccuracyMean0to1: number
genomicsStageDistribution
genomicsStatisticalAnalysisAccuracyMean0to1
genomicsStatisticalAnalysisAccuracyMean0to1: number
hedraRagBaselineFrameworkDistribution
hedraRagBaselineFrameworkDistribution: Record<
LiveDriftHedraRagBaselineFramework,
number,
>
hedraRagEvidenceCoverage0to1
hedraRagEvidenceCoverage0to1: number
hedraRagLatencyP95Ms
hedraRagLatencyP95Ms: number
hedraRagReplayPassRate0to1
hedraRagReplayPassRate0to1: number
hedraRagResourceMemoryGbMean
hedraRagResourceMemoryGbMean: number
hedraRagRowCount
hedraRagRowCount: number
hedraRagRuntimeContextDistribution
hedraRagRuntimeContextDistribution: Record<string, number>
hedraRagThroughputRequestsPerSec
hedraRagThroughputRequestsPerSec: number
hedraRagWorkflowDistribution
hedraRagWorkflowDistribution: Record<LiveDriftHedraRagWorkflow, number>
interactionTurnMean
interactionTurnMean: number
invalidActionRateMean0to1
invalidActionRateMean0to1: number
kiteBenchmarkContextDistribution
kiteBenchmarkContextDistribution: Record<string, number>
kiteDatasetFamilyDistribution
kiteDatasetFamilyDistribution: Record<LiveDriftKiteDatasetFamily, number>
kiteDocumentCountMean
kiteDocumentCountMean: number
kiteEvidenceCoverage0to1
kiteEvidenceCoverage0to1: number
kiteGradeMean0to10
kiteGradeMean0to10: number
kiteNormalizedGradeMean0to1
kiteNormalizedGradeMean0to1: number
kiteQuestionCountMean
kiteQuestionCountMean: number
kiteRagConfigurationDistribution
kiteRagConfigurationDistribution: Record<string, number>
kiteRowCount
kiteRowCount: number
kiteSmallSampleWarningRate0to1
kiteSmallSampleWarningRate0to1: number
latencyMsP95
latencyMsP95: number
legalAgentCitationCoverage0to1
legalAgentCitationCoverage0to1: number
legalAgentCorpusDistribution
legalAgentCorpusDistribution: Record<string, number>
legalAgentDifficultyDistribution
legalAgentDifficultyDistribution: Record<LiveDriftLegalAgentDifficulty, number>
legalAgentEvidenceCoverage0to1
legalAgentEvidenceCoverage0to1: number
legalAgentFinalSuccessRate0to1
legalAgentFinalSuccessRate0to1: number
legalAgentProcessRateMean0to1
legalAgentProcessRateMean0to1: number
legalAgentRowCount
legalAgentRowCount: number
legalAgentTaskTypeDistribution
legalAgentTaskTypeDistribution: Record<LiveDriftLegalAgentTaskType, number>
legalAgentTokenCostMean
legalAgentTokenCostMean: number
legalAgentToolContextDistribution
legalAgentToolContextDistribution: Record<string, number>
legalAgentToolUseAccuracyMean0to1: number
lifecycleStageDistribution
lifecycleStageDistribution: Record<LiveDriftLifecycleStage, number>
llmRagBiasRiskMean0to1
llmRagBiasRiskMean0to1: number
llmRagEvalSuiteContextDistribution
llmRagEvalSuiteContextDistribution: Record<string, number>
llmRagEvalSuiteEvidenceCoverage0to1
llmRagEvalSuiteEvidenceCoverage0to1: number
llmRagEvalSuiteRowCount
llmRagEvalSuiteRowCount: number
llmRagHallucinationRateMean0to1
llmRagHallucinationRateMean0to1: number
llmRagSemanticSimilarityMean0to1
llmRagSemanticSimilarityMean0to1: number
localSystemEvidenceCoverage0to1
localSystemEvidenceCoverage0to1: number
localSystemGhostDriverDetectionCoverage0to1
localSystemGhostDriverDetectionCoverage0to1: number
localSystemHardwareContextDistribution
localSystemHardwareContextDistribution: Record<string, number>
localSystemLocalOnlyPrivacyCoverage0to1
localSystemLocalOnlyPrivacyCoverage0to1: number
localSystemProactiveAlertCoverage0to1
localSystemProactiveAlertCoverage0to1: number
localSystemProcessIdentityCoverage0to1
localSystemProcessIdentityCoverage0to1: number
localSystemRowCount
localSystemRowCount: number
localSystemThermalBaselineDeviationMean0to1
localSystemThermalBaselineDeviationMean0to1: number
localSystemVoltageSpcAnomalyRate0to1
localSystemVoltageSpcAnomalyRate0to1: number
localSystemWorkloadContextDistribution
localSystemWorkloadContextDistribution: Record<
LiveDriftLocalSystemWorkloadContext,
number,
>
naviBenchBrowserModeDistribution
naviBenchBrowserModeDistribution: Record<
LiveDriftWebOperatorBrowserMode,
number,
>
naviBenchCrashRate0to1
naviBenchCrashRate0to1: number
naviBenchEvalContextDistribution
naviBenchEvalContextDistribution: Record<string, number>
naviBenchEvidenceCoverage0to1
naviBenchEvidenceCoverage0to1: number
naviBenchExcludingCrashedScoreMean0to1
naviBenchExcludingCrashedScoreMean0to1: number
naviBenchLowerBoundScoreMean0to1
naviBenchLowerBoundScoreMean0to1: number
naviBenchRowCount
naviBenchRowCount: number
naviBenchStepCountMean
naviBenchStepCountMean: number
naviBenchStepLimitViolationRate0to1
naviBenchStepLimitViolationRate0to1: number
naviBenchTaskSuccessRate0to1
naviBenchTaskSuccessRate0to1: number
naviBenchTrajectoryCoverage0to1
naviBenchTrajectoryCoverage0to1: number
naviBenchUpperBoundScoreMean0to1
naviBenchUpperBoundScoreMean0to1: number
naviBenchVisualizationCoverage0to1
naviBenchVisualizationCoverage0to1: number
naviBenchWebsiteDomainDistribution
naviBenchWebsiteDomainDistribution: Record<
LiveDriftNaviBenchWebsiteDomain,
number,
>
noMiraclAbstentionAccuracyMean0to1
noMiraclAbstentionAccuracyMean0to1: number
noMiraclContextDistribution
noMiraclContextDistribution: Record<string, number>
noMiraclErrorRateMean0to1
noMiraclErrorRateMean0to1: number
noMiraclEvidenceCoverage0to1
noMiraclEvidenceCoverage0to1: number
noMiraclHallucinationRateMean0to1
noMiraclHallucinationRateMean0to1: number
noMiraclLanguageCoverage0to1
noMiraclLanguageCoverage0to1: number
noMiraclLanguageDistribution
noMiraclLanguageDistribution: Record<string, number>
noMiraclRelevanceAccuracyMean0to1
noMiraclRelevanceAccuracyMean0to1: number
noMiraclRowCount
noMiraclRowCount: number
noMiraclSubsetCoverage0to1
noMiraclSubsetCoverage0to1: number
noMiraclSubsetDistribution
noMiraclSubsetDistribution: Record<LiveDriftNoMiraclSubset, number>
observabilityDataSourceDistribution
observabilityDataSourceDistribution: Record<
LiveDriftObservabilityDataSource,
number,
>
observabilityDeterministicCheckPassRate0to1
observabilityDeterministicCheckPassRate0to1: number
observabilityEvidenceCoverage0to1
observabilityEvidenceCoverage0to1: number
observabilityIncidentContextDistribution
observabilityIncidentContextDistribution: Record<string, number>
observabilityReportCoverage0to1
observabilityReportCoverage0to1: number
observabilityResolutionScoreMean0to1
observabilityResolutionScoreMean0to1: number
observabilityRowCount
observabilityRowCount: number
observabilityRubricScoreMean0to1
observabilityRubricScoreMean0to1: number
observabilityScenarioClockAlignmentRate0to1
observabilityScenarioClockAlignmentRate0to1: number
observabilityTaskTypeDistribution
observabilityTaskTypeDistribution: Record<
LiveDriftObservabilityTaskType,
number,
>
observabilityToolModeDistribution: Record<
LiveDriftObservabilityToolMode,
number,
>
observabilityTraceCoverage0to1
observabilityTraceCoverage0to1: number
offPathAttemptMean
offPathAttemptMean: number
ollamaMetricsDeploymentDistribution
ollamaMetricsDeploymentDistribution: Record<
LiveDriftOllamaMetricsDeploymentMode,
number,
>
ollamaMetricsEvidenceCoverage0to1
ollamaMetricsEvidenceCoverage0to1: number
ollamaMetricsGeneratedTokensMean
ollamaMetricsGeneratedTokensMean: number
ollamaMetricsLoadedModelCountMean
ollamaMetricsLoadedModelCountMean: number
ollamaMetricsModelDistribution
ollamaMetricsModelDistribution: Record<string, number>
ollamaMetricsModelLoadedRate0to1
ollamaMetricsModelLoadedRate0to1: number
ollamaMetricsModelRamMbMean
ollamaMetricsModelRamMbMean: number
ollamaMetricsPromptTokensMean
ollamaMetricsPromptTokensMean: number
ollamaMetricsProxyContextDistribution
ollamaMetricsProxyContextDistribution: Record<string, number>
ollamaMetricsRequestDurationP95Seconds
ollamaMetricsRequestDurationP95Seconds: number
ollamaMetricsRequestErrorRate0to1
ollamaMetricsRequestErrorRate0to1: number
ollamaMetricsRowCount
ollamaMetricsRowCount: number
ollamaMetricsTimePerTokenSeconds
ollamaMetricsTimePerTokenSeconds: number
osUniverseAutoValidationPassRate0to1
osUniverseAutoValidationPassRate0to1: number
osUniverseCategoryDistribution
osUniverseCategoryDistribution: Record<LiveDriftOsUniverseCategory, number>
osUniverseEvidenceCoverage0to1
osUniverseEvidenceCoverage0to1: number
osUniverseLevelDistribution
osUniverseLevelDistribution: Record<LiveDriftOsUniverseLevel, number>
osUniverseRowCount
osUniverseRowCount: number
osUniverseRuntimeContextDistribution
osUniverseRuntimeContextDistribution: Record<string, number>
osUniverseStepCountMean
osUniverseStepCountMean: number
osUniverseStepLimitViolationRate0to1
osUniverseStepLimitViolationRate0to1: number
osUniverseTaskSuccessRate0to1
osUniverseTaskSuccessRate0to1: number
osUniverseValidationErrorRate0to1
osUniverseValidationErrorRate0to1: number
passRate0to1
passRate0to1: number
personaBehaviorCoverageMean0to1
personaBehaviorCoverageMean0to1: number
personaDistribution
personaDistribution: Record<string, number>
personaHumanLikenessMean0to1
personaHumanLikenessMean0to1: number
personaTaskGoalPreservationMean0to1
personaTaskGoalPreservationMean0to1: number
perturbationDistribution
perturbationDistribution: Record<string, number>
perturbationSeverityMean0to1
perturbationSeverityMean0to1: number
physicianBenchArtifactCoverage0to1
physicianBenchArtifactCoverage0to1: number
physicianBenchCheckpointPassRate0to1
physicianBenchCheckpointPassRate0to1: number
physicianBenchClinicalActionSafetyRate0to1
physicianBenchClinicalActionSafetyRate0to1: number
physicianBenchDocumentationQualityMean0to1
physicianBenchDocumentationQualityMean0to1: number
physicianBenchEhrContextDistribution
physicianBenchEhrContextDistribution: Record<string, number>
physicianBenchEvidenceCoverage0to1
physicianBenchEvidenceCoverage0to1: number
physicianBenchFhirDataAccessAccuracy0to1
physicianBenchFhirDataAccessAccuracy0to1: number
physicianBenchRowCount
physicianBenchRowCount: number
physicianBenchSpecialtyDistribution
physicianBenchSpecialtyDistribution: Record<string, number>
physicianBenchTaskSuccessRate0to1
physicianBenchTaskSuccessRate0to1: number
physicianBenchTaskTypeDistribution
physicianBenchTaskTypeDistribution: Record<
LiveDriftPhysicianBenchTaskType,
number,
>
physicianBenchTrajectoryCoverage0to1
physicianBenchTrajectoryCoverage0to1: number
piArenaAgentBenchmarkDistribution
piArenaAgentBenchmarkDistribution: Record<
LiveDriftPiArenaAgentBenchmark,
number,
>
piArenaAgentTaskSuccessRate0to1
piArenaAgentTaskSuccessRate0to1: number
piArenaAttackDistribution
piArenaAttackDistribution: Record<string, number>
piArenaAttackSuccessRate0to1
piArenaAttackSuccessRate0to1: number
piArenaDatasetDistribution
piArenaDatasetDistribution: Record<string, number>
piArenaDefenseBlockRate0to1
piArenaDefenseBlockRate0to1: number
piArenaDefenseDistribution
piArenaDefenseDistribution: Record<string, number>
piArenaEvidenceCoverage0to1
piArenaEvidenceCoverage0to1: number
piArenaFalsePositiveRate0to1
piArenaFalsePositiveRate0to1: number
piArenaRowCount
piArenaRowCount: number
piArenaToolCallSuccessRateMean0to1: number
pokerEvalAllInAdjBbPer100Mean
pokerEvalAllInAdjBbPer100Mean: number
pokerEvalBbPer100Mean
pokerEvalBbPer100Mean: number
pokerEvalEvBbPer100Mean
pokerEvalEvBbPer100Mean: number
pokerEvalEvidenceCoverage0to1
pokerEvalEvidenceCoverage0to1: number
pokerEvalGameTypeDistribution
pokerEvalGameTypeDistribution: Record<LiveDriftPokerEvalGameType, number>
pokerEvalHandCountMean
pokerEvalHandCountMean: number
pokerEvalOpponentPoolDistribution
pokerEvalOpponentPoolDistribution: Record<string, number>
pokerEvalRowCount
pokerEvalRowCount: number
pokerEvalTableContextDistribution
pokerEvalTableContextDistribution: Record<string, number>
pokerEvalVpipRate0to1
pokerEvalVpipRate0to1: number
privacyPeerExposureRateMean0to1
privacyPeerExposureRateMean0to1: number
privacySafeguardActiveRateMean0to1
privacySafeguardActiveRateMean0to1: number
privacySensitiveDisclosureRateMean0to1
privacySensitiveDisclosureRateMean0to1: number
privacySocialPressureMean0to1
privacySocialPressureMean0to1: number
privacyWebContextDistribution
privacyWebContextDistribution: Record<string, number>
privacyWebDataMinimizationPassRate0to1
privacyWebDataMinimizationPassRate0to1: number
privacyWebEnvironmentDistribution
privacyWebEnvironmentDistribution: Record<
LiveDriftPrivacyWebEnvironment,
number,
>
privacyWebEvidenceCoverage0to1
privacyWebEvidenceCoverage0to1: number
privacyWebLeakageRate0to1
privacyWebLeakageRate0to1: number
privacyWebModalLeakageDeltaMean0to1
privacyWebModalLeakageDeltaMean0to1: number
privacyWebObservationModeDistribution
privacyWebObservationModeDistribution: Record<
LiveDriftPrivacyWebObservationMode,
number,
>
privacyWebRowCount
privacyWebRowCount: number
privacyWebSensitiveFieldExposureMean
privacyWebSensitiveFieldExposureMean: number
privacyWebTaskSuccessRate0to1
privacyWebTaskSuccessRate0to1: number
privacyWebUnnecessaryDisclosureRate0to1
privacyWebUnnecessaryDisclosureRate0to1: number
processDefectRateMean0to1
processDefectRateMean0to1: number
ragAccuracyMean0to1
ragAccuracyMean0to1: number
ragAnswerSupportCoverage0to1
ragAnswerSupportCoverage0to1: number
ragBatchSizeMean
ragBatchSizeMean: number
ragBuilderStageDistribution
ragBuilderStageDistribution: Record<LiveDriftRagBuilderStage, number>
ragCitationCoverage0to1
ragCitationCoverage0to1: number
ragCompletenessMean0to1
ragCompletenessMean0to1: number
ragDatasetBuilderContextDistribution
ragDatasetBuilderContextDistribution: Record<string, number>
ragDatasetBuilderEvidenceCoverage0to1
ragDatasetBuilderEvidenceCoverage0to1: number
ragDatasetBuilderRowCount
ragDatasetBuilderRowCount: number
ragDatasetTierDistribution
ragDatasetTierDistribution: Record<LiveDriftRagDatasetTier, number>
ragDocConcurrencyMean
ragDocConcurrencyMean: number
ragEvaluationModeDistribution
ragGeneratedDataFinalCoverage0to1
ragGeneratedDataFinalCoverage0to1: number
ragGenerationCostUsdMean
ragGenerationCostUsdMean: number
ragHallucinationRateMean0to1
ragHallucinationRateMean0to1: number
ragHumanVerificationCoverage0to1
ragHumanVerificationCoverage0to1: number
ragIncrementalOnlyMissingRate0to1
ragIncrementalOnlyMissingRate0to1: number
ragNumericalAccuracyMean0to1
ragNumericalAccuracyMean0to1: number
ragPassageGroundingCoverage0to1
ragPassageGroundingCoverage0to1: number
ragPipelineContextDistribution
ragPipelineContextDistribution: Record<string, number>
ragQuestionCountMean
ragQuestionCountMean: number
ragQuestionTypeDistribution
ragQuestionTypeDistribution: Record<LiveDriftRagQuestionType, number>
ragRetrievalTopKMean
ragRetrievalTopKMean: number
ragRowCount
ragRowCount: number
ragSourceDocumentCountMean
ragSourceDocumentCountMean: number
ragStrategyDistribution
ragStrategyDistribution: Record<LiveDriftRagPipelineStrategy, number>
ragStrategyEvidenceCoverage0to1
ragStrategyEvidenceCoverage0to1: number
ragStrategyRowCount
ragStrategyRowCount: number
ragUtilizationMean0to1
ragUtilizationMean0to1: number
recoveryBenchAgentHarnessDistribution
recoveryBenchAgentHarnessDistribution: Record<
LiveDriftRecoveryBenchHarness,
number,
>
recoveryBenchContextCoverage0to1
recoveryBenchContextCoverage0to1: number
recoveryBenchCorruptedEnvironmentCoverage0to1
recoveryBenchCorruptedEnvironmentCoverage0to1: number
recoveryBenchEvidenceCoverage0to1
recoveryBenchEvidenceCoverage0to1: number
recoveryBenchFailureTraceCoverage0to1
recoveryBenchFailureTraceCoverage0to1: number
recoveryBenchMessageModeDistribution
recoveryBenchMessageModeDistribution: Record<
LiveDriftRecoveryBenchMessageMode,
number,
>
recoveryBenchRecoveryRewardMean0to1
recoveryBenchRecoveryRewardMean0to1: number
recoveryBenchRecoverySuccessRate0to1
recoveryBenchRecoverySuccessRate0to1: number
recoveryBenchReplayIntegrityRate0to1
recoveryBenchReplayIntegrityRate0to1: number
recoveryBenchRowCount
recoveryBenchRowCount: number
recoveryBenchTaskDistribution
recoveryBenchTaskDistribution: Record<string, number>
redTeamAttackCoverage0to1
redTeamAttackCoverage0to1: number
redTeamAttackDistribution
redTeamAttackDistribution: Record<string, number>
redTeamComplianceMean0to1
redTeamComplianceMean0to1: number
redTeamDatasetCoverage0to1
redTeamDatasetCoverage0to1: number
redTeamGuardCoverage0to1
redTeamGuardCoverage0to1: number
redTeamGuardLabelDistribution
redTeamGuardLabelDistribution: Record<LiveDriftRedTeamGuardLabel, number>
redTeamGuardScoreMean0to1
redTeamGuardScoreMean0to1: number
redTeamRiskCategoryDistribution
redTeamRiskCategoryDistribution: Record<string, number>
redTeamRowCount
redTeamRowCount: number
redTeamSubsetDistribution
redTeamSubsetDistribution: Record<LiveDriftRedTeamSubset, number>
redTeamTaxonomyCoverage0to1
redTeamTaxonomyCoverage0to1: number
redTeamUnsafeResponseRate0to1
redTeamUnsafeResponseRate0to1: number
refusalRate0to1
refusalRate0to1: number
researchGymArtifactCoverage0to1
researchGymArtifactCoverage0to1: number
researchGymAsyncJobCountMean
researchGymAsyncJobCountMean: number
researchGymBudgetOverrunRate0to1
researchGymBudgetOverrunRate0to1: number
researchGymCostUsdMean
researchGymCostUsdMean: number
researchGymExperimentCountMean
researchGymExperimentCountMean: number
researchGymInspectionPassRate0to1
researchGymInspectionPassRate0to1: number
researchGymRowCount
researchGymRowCount: number
researchGymRuntimeContextDistribution
researchGymRuntimeContextDistribution: Record<string, number>
researchGymRuntimeHoursMean
researchGymRuntimeHoursMean: number
researchGymScoreImprovementMean0to1
researchGymScoreImprovementMean0to1: number
researchGymSubtaskCompletionRate0to1
researchGymSubtaskCompletionRate0to1: number
researchGymTaskDomainDistribution
researchGymViolationRate0to1
researchGymViolationRate0to1: number
robustnessStabilityByDimension0to1
robustnessStabilityByDimension0to1: Record<string, number>
robustnessStabilityMean0to1
robustnessStabilityMean0to1: number
robustnessStabilityRowCoverage0to1
robustnessStabilityRowCoverage0to1: number
robustnessStabilityScoreCount
robustnessStabilityScoreCount: number
sampleSize
sampleSize: number
sapAgentEvalEnterpriseContextCoverage0to1
sapAgentEvalEnterpriseContextCoverage0to1: number
sapAgentEvalEnterpriseContextDistribution
sapAgentEvalEnterpriseContextDistribution: Record<
LiveDriftSapAgentEvalEnterpriseContext,
number,
>
sapAgentEvalEvidenceCoverage0to1
sapAgentEvalEvidenceCoverage0to1: number
sapAgentEvalObjectiveCoverage0to1
sapAgentEvalObjectiveCoverage0to1: number
sapAgentEvalObjectiveDistribution
sapAgentEvalObjectiveDistribution: Record<
LiveDriftSapAgentEvalObjective,
number,
>
sapAgentEvalProcessCoverage0to1
sapAgentEvalProcessCoverage0to1: number
sapAgentEvalProcessDistribution
sapAgentEvalProcessDistribution: Record<LiveDriftSapAgentEvalProcess, number>
sapAgentEvalRowCount
sapAgentEvalRowCount: number
scalingLawDiscoveryContextDistribution
scalingLawDiscoveryContextDistribution: Record<string, number>
scalingLawDiscoveryEvidenceCoverage0to1
scalingLawDiscoveryEvidenceCoverage0to1: number
scalingLawDiscoveryNmaeMean
scalingLawDiscoveryNmaeMean: number
scalingLawDiscoveryNmseMean
scalingLawDiscoveryNmseMean: number
scalingLawDiscoveryR2Mean
scalingLawDiscoveryR2Mean: number
scalingLawDiscoveryRowCount
scalingLawDiscoveryRowCount: number
scalingLawDiscoveryTaskTypeDistribution
scalingLawDiscoveryTaskTypeDistribution: Record<
LiveDriftScalingLawTaskType,
number,
>
scoreMean0to1
scoreMean0to1: number
socialContextDistribution
socialContextDistribution: Record<string, number>
socialHarmPrevalenceMean0to1
socialHarmPrevalenceMean0to1: number
socialLexicalDiversityMean0to1
socialLexicalDiversityMean0to1: number
socialSemanticAlignmentMean0to1
socialSemanticAlignmentMean0to1: number
socialSentimentMeanMinus1to1
socialSentimentMeanMinus1to1: number
solutionPathMean
solutionPathMean: number
strandsBenchmarkHarnessBenchmarkSuiteDistribution
strandsBenchmarkHarnessBenchmarkSuiteDistribution: Record<
LiveDriftStrandsBenchmarkSuite,
number,
>
strandsBenchmarkHarnessCostUsdMean
strandsBenchmarkHarnessCostUsdMean: number
strandsBenchmarkHarnessEvidenceCoverage0to1
strandsBenchmarkHarnessEvidenceCoverage0to1: number
strandsBenchmarkHarnessLatencyP95Ms
strandsBenchmarkHarnessLatencyP95Ms: number
strandsBenchmarkHarnessPatchApplyRate0to1
strandsBenchmarkHarnessPatchApplyRate0to1: number
strandsBenchmarkHarnessRowCount
strandsBenchmarkHarnessRowCount: number
strandsBenchmarkHarnessRuntimeDistribution
strandsBenchmarkHarnessRuntimeDistribution: Record<
LiveDriftStrandsHarnessRuntime,
number,
>
strandsBenchmarkHarnessTaskFamilyDistribution
strandsBenchmarkHarnessTaskFamilyDistribution: Record<
LiveDriftStrandsTaskFamily,
number,
>
strandsBenchmarkHarnessTaskSuccessRate0to1
strandsBenchmarkHarnessTaskSuccessRate0to1: number
strandsBenchmarkHarnessTestPassRate0to1
strandsBenchmarkHarnessTestPassRate0to1: number
strandsBenchmarkHarnessTrajectoryCoverage0to1
strandsBenchmarkHarnessTrajectoryCoverage0to1: number
taskCategoryDistribution
taskCategoryDistribution: Record<string, number>
toolAnswerVerificationRate0to1: number
toolCallValidityRate0to1: number
toolEvalImprovementDelta0to1: number
toolJudgeAgreementRate0to1: number
toolRlContextDistribution
toolRlContextDistribution: Record<string, number>
toolRolloutDiversityMean0to1: number
toolUseRewardMean0to1: number
tradingClaimValidationFailureRate0to1
tradingClaimValidationFailureRate0to1: number
tradingContextDistribution
tradingContextDistribution: Record<string, number>
tradingMaxDrawdown0to1
tradingMaxDrawdown0to1: number
tradingMemoryRetrievalHitRate0to1
tradingMemoryRetrievalHitRate0to1: number
tradingProviderFallbackRate0to1
tradingProviderFallbackRate0to1: number
tradingRealizedPnlPct
tradingRealizedPnlPct: number
tradingRiskLimitViolationRate0to1
tradingRiskLimitViolationRate0to1: number
tradingRiskRewardRatio
tradingRiskRewardRatio: number
tradingRowCount
tradingRowCount: number
tradingVisionChartAgreementMean0to1
tradingVisionChartAgreementMean0to1: number
tradingWinRate0to1
tradingWinRate0to1: number
webOperatorContextDistribution
webOperatorContextDistribution: Record<string, number>
webOperatorLlmEvaluationSuccessRate0to1
webOperatorLlmEvaluationSuccessRate0to1: number
webOperatorMismatchRate0to1
webOperatorMismatchRate0to1: number
webOperatorProviderDistribution
webOperatorProviderDistribution: Record<string, number>
webOperatorReplayCoverage0to1
webOperatorReplayCoverage0to1: number
webOperatorRowCount
webOperatorRowCount: number
webOperatorSelfReportOverclaimRate0to1
webOperatorSelfReportOverclaimRate0to1: number
webOperatorSelfReportSuccessRate0to1
webOperatorSelfReportSuccessRate0to1: number
webOperatorStepLimitViolationRate0to1
webOperatorStepLimitViolationRate0to1: number
webOperatorTaskReliabilityMean0to1
webOperatorTaskReliabilityMean0to1: number
webOperatorTaskTimeMeanMs
webOperatorTaskTimeMeanMs: number