dubai-02-wave2: the pre-registered statistic before R6 and R7, as published, and on wave 1's 15 questions

before R6 and R7 (gate.py, 23 September, 10:47 UTC): 40 questions, 105 of 115 pairs below the mean (91.3%), median ratio 0.653, 5 excluded (G4 chatgpt-gemini, G4 chatgpt-ai_mode, G4 gemini-ai_mode, T3 chatgpt-ai_mode, T3 gemini-ai_mode); REPLICATES
as published (committed q1_surface_overlap.csv)     : 40 questions, 104 of 115 pairs below the mean (90.4%), median ratio 0.652, 5 excluded (G4 chatgpt-gemini, G4 chatgpt-ai_mode, G4 gemini-ai_mode, T3 chatgpt-ai_mode, T3 gemini-ai_mode); REPLICATES

as published, wave 1's 15 questions (not pre-registered): 15 questions, 37 of 42 pairs below the mean (88.1%), median ratio 0.642, 3 excluded (G4 chatgpt-gemini, G4 chatgpt-ai_mode, G4 gemini-ai_mode); REPLICATES
