eval-harness

reference

silver/orrery_triage_v1_labelled_outcomes silver

Promoted as silver: consistent, not a claim that any arm is right; good for ranking arms against each other.
Promoted from rule silver_orrery_triage_v1_labelled_outcomes, 2026-11-02, by owner: What production recorded afterwards ranks the arms beside the owner's gold; it never decides.

01what it is

tier
silver
dataset
orrery_triage_v1_labelled
promoted from
rule silver_orrery_triage_v1_labelled_outcomes
promoted by
owner
why
What production recorded afterwards ranks the arms beside the owner's gold; it never decides.
items
34
created
2026-11-02T08:17:27Z

02scored against it · 0

No arm is scored against it yet.