Calibration
MMLU-Redux → AA-Omniscience Accuracy
AA-Omniscience Accuracy is estimated from MMLU-Redux with a inverse Michaelis–Menten curve fitted on 7 models measured on both: y = 0.01423·(x − 0.0000) / (0.0000 + 0.9994 − x), R² = 0.95, cross-validated error 4.5 pp. It is used for 4 estimates.
| Estimated model | MMLU-Redux | AA-Omniscience Accuracy | Source |
|---|---|---|---|
| Mellum2-12B-A2.5B-Instruct | 78.1% | 5.1% | estimated ± 4.5 pp, low confidence |
| Mellum2-12B-A2.5B-Thinking | 86.2% | 8.9% | estimated ± 4.5 pp, medium confidence |
| MiniCPM5-1B | 70.1% | 3.3% | estimated ± 4.5 pp, low confidence |
| Ternary Bonsai 2 27B | 89.1% | 11.7% | estimated ± 4.5 pp, medium confidence |