| Benchmark | Model | Rank | Results |
|---|---|---|---|
| zero-shot-question-answering-on-sciq | MeKi (4B) | #7 | Accuracy: 87.2 |
| language-modelling-on-lambada | MeKi (4B) | #46 | Accuracy: 50.1 |
| question-answering-on-boolq | MeKi (4B) | #79 | Accuracy: 64.4 |
| common-sense-reasoning-on-arc-easy | MeKi (4B) | #91 | Accuracy: 42.2 |
| common-sense-reasoning-on-winogrande | MeKi (4B) | #107 | Accuracy: 61.6 |
| question-answering-on-piqa | MeKi (4B) | #111 | Accuracy: 75.4 |
| sentence-completion-on-hellaswag | MeKi (4B) | #140 | Accuracy: 62.3 |