Benchmark Results Dashboard

The benchmark harness runs five mitigation strategies across three model families on all seven audits, scored on six fairness metrics with a bootstrap confidence interval and a permutation-test p-value. This page explores those same numbers interactively - filter by audit, strategy, model, protected attribute, and metric instead of grepping CSVs.

🔒 Everything below runs locally in your browser - the bundled CSVs are fetched from this same site, and a dropped file never leaves your machine.