Recognises points out of proportion to the construct
Weights that do not reflect the construct: format worth as much as skill.
Can AI do this?
AI does this well Most of the tested AI models master this. 3 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 94% | Mastered |
| gemini-3.5-flash-lite | Low cost | 50% | Not shown |
| gpt-6-luna | Low cost | 94% | Mastered |
| claude-haiku-5-5 | Low cost | 94% | Mastered |
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: Weights following the construct
How it is assessed: Flags the deficient case and passes the acceptable one. Deficient: Q2 "valid YAML" 4 of 12 points Acceptable: Format at most about a tenth of the points
Also known as: Points out of proportion
Related knobits
Builds on
All 15 knobits of Write and review test items.
Source: Review checklist 02-test-items, row 7 (Standards 4)