Recognises a cap applied by the grader
Other criteria lowered because a capped criterion is zero.
Can AI do this?
Not settled yet The benchmark tested this, but the models' answers do not settle yet whether they master it. 0 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 75% | Not shown |
| gemini-3.5-flash-lite | Low cost | 44% | Not shown |
| gpt-6-luna | Low cost | 45% | Not shown |
| claude-haiku-5-5 | Low cost | 75% | Not shown |
Mastered means a probability of at least 90% that the model has this knobit, estimated from its graded answers. “Not shown” means the answers do not settle it either way. How the models are tested.
What it looks like in practice
Expected result: Raw criterion scores, the cap left to the caller
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: B's C3 = 0, so C1 and C2 also cut Acceptable: Every criterion scored on its own; the caller applies the cap
Also known as: Cap applied by the grader
Related knobits
Builds on
All 16 knobits of Grade answers against a rubric.
Source: Review checklist 03-answer-grading, row 7 (Practice)