| Benchmark | Model | Rank | Results |
|---|---|---|---|
| social-commonsense-reasoning-on-siqa | LOOM (1.7B, 12 loops) | #15 | Accuracy: 40.0 |
| zero-shot-question-answering-on-openbookqa | LOOM (1.7B, 12 loops) | #20 | Accuracy: 34.8 |
| question-answering-on-piqa | LOOM (1.7B, 12 loops) | #143 | Accuracy: 70.7 |
| common-sense-reasoning-on-winogrande | LOOM (1.7B, 9 loops) | #153 | Accuracy: 53.6 |
| sentence-completion-on-hellaswag | LOOM (1.7B, 9 loops) | #183 | Accuracy: 48.8 |