Dashboard / World Model🟡 Sample Data
Forecasting with a living world model
The world model combines observations, memory, and simulation to estimate current state and predict futures. Causal graphs power counterfactual reasoning.
—
Causal Pairs
Observed cause-effect
—
Drift Score
Parameter divergence
∞
Counterfactuals
Replay any scenario
99.8%
Retention Rate
Memory stability
Causal Graph
Start the Synaps AGI engine to view live causal data.
Counterfactual Replay
Ask "what if X hadn't happened?" and the causal engine replays the scenario without that event. Lift-based ranking surfaces the strongest causal links.
- Event-pair counter tracks co-occurrence
- Lift metric separates correlation from causation
- Predict effects before executing interventions
Drift Detection
Continuous monitoring of parameter drift across memory tiers. When drift exceeds thresholds, the system triggers automatic consolidation.
Engine offline — drift monitoring requires a running Synaps AGI instance.