teamchong 455ab636a2 eval(patch-probe): billing grid + OCR phase-sweep findings ...

Probe 1: both fable-5 and sonnet-5 bill vision on a 28x28 patch grid
(image_tokens = 3 + ceil(W/28)*ceil(H/28)); snap rows≡6 (mod 7),
cols≡4 (mod 28) to avoid stranding paid patch area.

Probe 2: patch-boundary straddling does NOT affect OCR accuracy
(cols: 4.47% vs 4.60%; rows: z=-0.45, 7-offset paired sweep with
line fixed effects). Real misread drivers: high-entropy runs
(bimodal derailment on base64 blobs), 5x8 confusables (w/W, 8/0),
line wraps. Harnesses: count-tokens-sweep, accuracy-phase-probe
(line-DP-aligned scoring), rescore-sweep (offline paired analysis).

2026-07-12 18:59:31 -04:00

..

fable-H-w56.csv

eval(patch-probe): billing grid + OCR phase-sweep findings

2026-07-12 18:59:31 -04:00

fable-W-h56.csv

eval(patch-probe): billing grid + OCR phase-sweep findings

2026-07-12 18:59:31 -04:00

sonnet-H-w56.csv

eval(patch-probe): billing grid + OCR phase-sweep findings

2026-07-12 18:59:31 -04:00

sonnet-W-h56.csv

eval(patch-probe): billing grid + OCR phase-sweep findings

2026-07-12 18:59:31 -04:00