DEBUG.EVAL · GALLERY
Debug Eval
Per-entry diagnostic. POST /api/prompt-tuner/batch-eval with knobs (crop, severity, custom prompt).
Cards compare model verdicts to CD ground truth.
CACHE 0
▢
NO ENTRIES MATCH CURRENT FILTER
Per-entry diagnostic. POST /api/prompt-tuner/batch-eval with knobs (crop, severity, custom prompt).
Cards compare model verdicts to CD ground truth.