| Benchmark | Model | Rank | Results |
|---|---|---|---|
| zero-shot-question-answering-on-sciq | Parallax (1.7B) (LM Evaluation Harness) | #5 | Accuracy: 88.7 |
| common-sense-reasoning-on-arc-easy | Parallax (1.7B) | #43 | Accuracy: 73.27 |
| language-modelling-on-lambada | Parallax (1.7B) | #45 | Accuracy: 50.26Perplexity: 10.8 |
| question-answering-on-boolq | Parallax (1.7B) | #78 | Accuracy: 64.59 |
| question-answering-on-piqa | Parallax (1.7B) | #101 | Accuracy: 76.39 |
| common-sense-reasoning-on-winogrande | Parallax (1.7B) | #112 | Accuracy: 60.77 |
| sentence-completion-on-hellaswag | Parallax (1.7B) | #133 | Accuracy: 64.54 |