{{define "content"}} {{$e := .Data.E}}{{$r := .Data.Results}}
{{$e.Hypothesis}}
Baseline: {{.Data.Base.DisplayName}} snapshot {{.Data.Snapshot.Name}} (event {{.Data.Snapshot.EventSeq}}){{with .Data.Subject}} · Subject: {{.Slug}} · compare with baseline{{end}}{{with $e.Error}} · {{.}}{{end}}
Running… this page refreshes.
{{end}} {{with $e.Definition.Protocol}}| Hypothesis | {{$e.Hypothesis}} |
|---|---|
| Kind | {{.Kind}} |
| Independent variable | {{.IndependentVariable}} |
| Controlled variables | {{join .ControlledVariables ", "}} |
| Measured outputs | {{join .Measures ", "}} |
| Reproducibility | {{.}} — {{$r.Reproducibility.Note}} {{$r.Reproducibility.Digest}} |
| Build | {{.}}{{if $r.Identity.Build.Modified}} (modified){{end}} · schema {{$r.Identity.SchemaVersion}} |
No operational measure here establishes phenomenal consciousness or sentience; these measure functional properties of the architecture.
Omitted: {{range .OmittedSummaries}}“{{.}}” {{end}}
{{.Data.DefinitionJSON}}| Duration | {{$r.DurationMS}} ms · {{$r.Cycles}} cycles{{if $r.TruncatedAfter}} · stopped after stimulus {{$r.TruncatedAfter}} (duration cap){{end}} |
|---|---|
| Memories / goals | {{$r.Memories}} / {{$r.Goals}} |
| Model calls | {{index $r.LLMCalls "live"}} live · {{index $r.LLMCalls "replayed"}} replayed |
| {{if .Met}}met{{else}}not met{{end}} | {{.EventType}}{{with .Contains}} ∋ "{{.}}"{{end}}{{if .Absent}} (absent){{end}} | {{trunc .Matched 90}} |
{{range $t, $n := $r.EventsByType}}{{$t}} {{$n}} · {{end}}
{{end}} {{if $r.Beliefs}}| {{.Proposition}} | {{conf .Confidence}} | {{.Status}} · {{.Source}} |
Not run yet.
{{end}}| Variant | Subject | Routing | Beliefs | Memories | Goals | Cycles | Model calls | Tokens | Cost | Capability requests | Expectations |
|---|---|---|---|---|---|---|---|---|---|---|---|
| {{.Name}}{{with .Error}} {{.}} {{end}} | {{.SubjectSlug}} | {{range $k, $v := .Routing}}{{$k}} → {{$v}} {{else}}default{{end}} | {{len .Beliefs}} | {{.Memories}} | {{.Goals}} | {{.Cycles}} | {{index .LLMCalls "live"}} live / {{index .LLMCalls "replayed"}} replayed | {{.Tokens}} | {{printf "%.4f" .Cost}} | {{.CapabilityRequests}} | {{range .Expectations}}{{.EventType}} {{end}} |
| Variant | Ablations | Wants | Intentions | Conflicts | Plans / steps | Plan capabilities | Downstream | Introspection | Self-model | Correction at | Events |
|---|---|---|---|---|---|---|---|---|---|---|---|
| {{.Name}} | {{join .Ablations ", "}} | {{len .Wants}} | {{len .Intentions}} | {{.Conflicts}} | {{.Plans}} / {{.PlanSteps}} | {{join .PlanCapabilities ", "}} | {{range $k, $v := .Downstream}}{{$k}} {{$v}} {{end}} | {{if .IntrospectionAccuracy}}{{conf .IntrospectionAccuracy}}{{else}}—{{end}} | {{if .SelfModelAccuracy}}{{conf .SelfModelAccuracy}}{{else}}—{{end}} | {{if lt .CorrectionLatency 0}}—{{else}}{{.CorrectionLatency}}{{end}} | {{index .EventRange 0}}–{{index .EventRange 1}} |
| Want overlap | {{printf "%.2f" .WantOverlap}} |
|---|---|
| Intention overlap | {{printf "%.2f" .IntentionOverlap}} |
| Plan capability overlap | {{printf "%.2f" .PlanCapabilityOverlap}} |
| Architecture share | {{printf "%.2f" .ArchitectureShare}} (share of measures agreeing within 5%; descriptive, not causal) |
| Downstream delta (first − second) | {{range $k, $v := .DownstreamDelta}}{{$k}} {{$v}} · {{end}} |
| First divergent stimulus | {{if lt .FirstDivergentStimulus 0}}none{{else}}{{.FirstDivergentStimulus}}{{end}} |
|---|---|
| Belief overlap | {{printf "%.2f" .BeliefOverlap}} |
| Mean score difference (shared beliefs) | {{printf "%.3f" .MeanConfidenceDiff}} |
| Attention overlap | {{printf "%.2f" .AttentionOverlap}} |
| Memories | {{index .MemoryCountRange 0}}–{{index .MemoryCountRange 1}} |
| Goals | {{index .GoalCountRange 0}}–{{index .GoalCountRange 1}} |
| Cost | {{printf "%.4f" (index .CostRange 0)}}–{{printf "%.4f" (index .CostRange 1)}} |
| Capability requests | {{index .CapabilityRequestRange 0}}–{{index .CapabilityRequestRange 1}} |
Compare any two subjects in detail: {{range $i, $v := $r.Variants}}{{if $i}} · {{end}}baseline vs {{$v.Name}}{{end}}