Finite-Time Lyapunov Exponents (FTLE)

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Fast Lyapunov Indicator (λF)method0.774
  Metric directly localizing saddle-like boundaries between solution routes by measuring maximal local trajectory separation.
• Pre-bifurcation, all finite-time Lyapunov exponents are negative, so no transient chaos occursfinding0.768
  Establishes the baseline (non-chaotic) regime before the bifurcation.
• Lyapunov Functionconcept0.724
  Variational free energy acts as Lyapunov function for neuronal dynamics, ensuring convergence.
• Fast Lyapunov indicator λF strongly correlates with the number of distinct solutions a reasoning trajectory passes through before convergingfinding0.710
  Links saddle-crossing intensity to the number of candidate-answer switches during reasoning.
• F-statistics and Linear Probes for Feature Selectionmethod0.696
  Method to select d_steer top-activated SAE features for constructing control vectors
• Only the multistep 'core' variables (not directly-substitutable variables) produce positive FTLE in the trained looped transformerfinding0.681
  Localizes transient chaos to the sub-algorithm requiring multi-step Gaussian elimination.
• Gradient Descent on Free Energymethod0.677
  Optimization procedure for simultaneously updating action selection and perception; uses step size ζ (default 4).
• Fluctuation Theoremconcept0.676
  Statistical mechanics principle governing entropy increase in open systems; free energy minimization resists this dispersal.