jevPOST
A small tool for picking thresholds on Jev scores
While building a distribution tool with Jev, the developer struggled with what threshold to ship at 0.67 and built a tool that tests a decision against self-labeled examples to show what each threshold catches and misses.
Yusuke
@yusukelpArchived 2026.09.28Originally posted 2026.09.27
Yusuke @yusukelpOriginally posted 2026.09.27
While building a distribution tool with Jev, I kept asking: “Jev says 0.67. What threshold should I actually ship?” So I built a small tool for that. Test a decision against examples you've labeled yourself, then see what each threshold catches and misses. Dropping a v0 tomorrow 👀
This post quotes another post: x.com/yusukelp/status/2101715040621715577
x.com/yusukelp/status/2104218730272260324Archived 2026.09.28 10:02