Skip to main content

Runs and Results

Purpose​

Use Results to understand what happened during a run and what needs to change next.

When to use it​

Open a run or result view after starting an eval or playground run.

Prerequisites​

  • You have started at least one run.

Steps​

  1. Open the run detail page from Evals, Playground, or the project Results area.
  2. Review the run status and latest step.
  3. Check how many results completed successfully or failed.
  4. Inspect per-row outputs, correctness, and any scoring details.
  5. Review console output or error details when present.
  6. Export results when you need a CSV copy for analysis outside Spec27.

Expected result​

You can explain what the run did, what passed or failed, and where to iterate next.

What to know​

  • Run detail pages can show output, status, correctness, and logs together.
  • Judge-based runs may include explanations and vote details.
  • Adversarial runs can surface adversarial inputs instead of only the seeded primary inputs.