The eval you actually need

The eval you actually need syllabus

A benchmark measures the average task. You have one task, and awkward inputs. This is how to measure that instead.

Module 0 — Why leaderboards lie

Not because they're dishonest, but because they answer a question that isn't yours.

  1. The average task is not your task — A number that moved three points on a benchmark may move zero on your problem. (2 min)
  2. Three failures a benchmark will never show you — Refusals, near-misses and confident nonsense, none of which move an accuracy number. (2 min)

Module 1 — Tasks as data

Adding a probe should be appending to a list, never editing a runner.

  1. One record per probe — The shape that lets you add the hundredth test case as cheaply as the second. (2 min)
  2. Scoring without fooling yourself — Exact match is too strict, a model judge is too kind, and both are measurable. (2 min)