🎓 Ages 15–18 · Grades 9–12 · Scientific Method & Evaluation

BenchForge

Every leaderboard, model card, and study is an argument — and the person who designed the test quietly decided who wins. Read a real evaluation, predict what the numbers will do, then watch a deterministic engine work it out. Learn to read the test, not just the score.

Start here

Concept kits — 16 kits of evaluation & scientific-method reasoning

From “what an evaluation IS” to leakage, hold-out splits, the sealed envelope, metric choice, and gaming a benchmark. Progress stays on this device.

Meet the cast — then teach one

Not mascots — everyday reasoners, each showing one idea in action. On the Teach page you help one of them design a test that can’t be fooled.

  • The Referee — runs the neutral, blind, prospective test (the mentor)
  • Splitter — separates train from test — and hunts leakage between them
  • Binner — stratifies cases by novelty so the hard band shows
  • Metric — picks the yardstick — and shows the winner change when it swaps

🎓 Open the case coach

How it works

BenchForge is about the reasoning, not memorizing terms, and a wrong prediction is just a place to learn. Every answer is worked out by a small deterministic engine, so the numbers are always exact. Your progress is saved only on this device: no accounts, nothing sent anywhere.

More ways to play