52 gate-passed questions · 48 cells · 14 occupied · go to this type's cells
Bins were cut on this corpus at 240×135@16 (CLEAR_PX 25, motion floor 19 px). Margin bins: a near-edge band [1, 1.25) above the gate boundary, then corpus quantiles; salience bins: a visibility floor, then quantiles.
0.25 / edge_frac (>=1 passes) · measurable on 52 questions
| margin bin | numeric range |
|---|---|
| near-edge [1,1.25) | [−∞, 1.25) |
| bin0 | [1.25, 30.95) |
| bin1 | [30.95, ∞) |
Ask: is this the same quantity the gate certifies on for this type? Are the states (e.g. 'same', 'gone') the right one-bin exceptions? Are the edges where a change in difficulty actually happens?
motion extent: screen path / W · floor 0.07917
| salience bin | numeric range |
|---|---|
| <floor (0.0792) | [−∞, 0.07917) |
| bin0 | [0.07917, 0.3484) |
| bin1 | [0.3484, 0.4729) |
| bin2 | [0.4729, ∞) |
Ask: does this axis capture what makes a question of this type perceptually hard, independently of the answer? Is the floor the right anti-gaming boundary?
answer (branch 1) · labels: the bottom edge, the left edge, the right edge, the top edge
Compatible margin bins per context label (cells outside these are definitionally impossible):
| answer | n |
|---|---|
| the right edge | 24 |
| the left edge | 23 |
| the bottom edge | 3 |
| the top edge | 2 |