add pass@k and pass^k metrics with configurable threshold#944

Merged

willccbb merged 6 commits intomainfrom

hallerite/passatk

Feb 24, 2026

Commits on Feb 20, 2026

Commits on Feb 22, 2026

Commits on Feb 23, 2026

docs: update reference.md with pass_at_k, pass_all_k, pass_threshold fields

cursoragent
and
willccbb
committed
Fix compute_pass_at_k: exclude incomplete example groups from metric

cursoragent
and
willccbb
committed