File size: 1,599 Bytes
92c59c2 c3a24db 92c59c2 c3a24db 92c59c2 c3a24db 92c59c2 c3a24db | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 | # P1 Fixture Sanity
This folder now contains three paired low-fidelity/high-fidelity `P1` fixtures:
- `boundary_default_reset.json`
- `bad_low_iota.json`
- `lowfi_feasible_local.json`
Calibration source:
- coarse measured sweep artifact: `baselines/sweep_results/measured_sweep_20260308T050043Z.json`
- targeted local probes around the default reset seed
Current interpretation:
- `boundary_default_reset.json`
- near-boundary reset reference
- infeasible, but close enough that short repair episodes are realistic
- `bad_low_iota.json`
- clearly bad infeasible case
- demonstrates that edge iota can be the blocking constraint even when triangularity is already acceptable
- `lowfi_feasible_local.json`
- low-fidelity feasible local target
- reachable from the default reset band with two intuitive knob increases
Observed from paired run:
- low-fi vs high-fi feasibility alignment and metric deltas are documented in `baselines/fixture_high_fidelity_pairs.json`.
- decision on whether any reset seed should be changed from the current default
Current paired summary (`baselines/fixture_high_fidelity_pairs.json`):
- `bad_low_iota.json`:
- both fidelities infeasible
- low/high feasibility match: `true`
- low/high score match: both `0.0`
- `boundary_default_reset.json`:
- both fidelities infeasible
- low/high feasibility match: `true`
- low/high score match: both `0.0`
- `lowfi_feasible_local.json`:
- both fidelities feasible
- low/high feasibility match: `true`
- high-fidelity score improved slightly: `0.29165951078327634` → `0.2920325118884466`
|