{ "seed": 2734, "derivative_verification": { "max_abs_slope_error": 1.4988010832439613e-13, "max_contract_residual": 0.0, "predicted_unit_rate": 1.0, "observed_normalized_slope": 1.0000000000001152 }, "math_verification": { "grid_points": 6480, "conservative_contract_violations": 0, "conservative_violation_rate": 0.0, "underestimated_gamma": 0.75, "underestimate_unsafe_onset_median_L": 1.08675, "expected_physical_boundary_median_L": 1.0867088607594937 }, "gamma_boundary_sweep": { "predicted_gamma_transition": 0.9090909090909091, "observed_gamma_transition_grid": 0.91, "transition_abs_error": 0.0009090909090909705, "unsafe_permissions_at_conservative_gamma": 0, "unsafe_permissions_at_underestimated_gamma": 51, "predicted_unsafe_onset_gamma": 0.9090909090909091, "observed_first_unsafe_gamma": 0.9 }, "evaluation_scaling": { "rows": [ { "gamma": 0.5, "budget": 1.9, "observed_evals": 53, "predicted_evals": 52.631578947368425, "rate_error": 0.006999999999999928 }, { "gamma": 1.0, "budget": 0.8999999999999999, "observed_evals": 112, "predicted_evals": 111.11111111111113, "rate_error": 0.007999999999999842 }, { "gamma": 1.5, "budget": 0.5666666666666665, "observed_evals": 177, "predicted_evals": 176.47058823529414, "rate_error": 0.002999999999999848 }, { "gamma": 2.0, "budget": 0.4, "observed_evals": 250, "predicted_evals": 250.0, "rate_error": 0.0 } ], "rate_ratio_observed_gamma2_over_gamma1": 2.232142857142857, "rate_ratio_predicted_gamma2_over_gamma1": 2.2499999999999996 }, "scheduler_comparison": { "episodes": 3000, "periodic_unsafe_handoff_rate": 0.288, "naive_event_unsafe_handoff_rate": 0.057666666666666665, "contract_unsafe_handoff_rate": 0.0, "contract_launch_fraction": 0.792, "contract_fallback_fraction": 0.208, "periodic_evaluations_per_episode": 1.0, "contract_evaluations_per_episode": 0.792 } }