| Benchmark | Model | Rank | Results |
|---|---|---|---|
| common-sense-reasoning-on-commonsenseqa | STaR (on GPT-J) | #16 | Accuracy: 72.3 |
| common-sense-reasoning-on-commonsenseqa | STaR without Rationalization (on GPT-J) | #18 | Accuracy: 68.8 |
| common-sense-reasoning-on-commonsenseqa | GPT-J Direct Finetuned | #26 | Accuracy: 60.0 |
| common-sense-reasoning-on-commonsenseqa | Few-shot CoT LaMDA 137B | #30 | Accuracy: 55.6 |
| common-sense-reasoning-on-commonsenseqa | Few-shot CoT GPT-J | #32 | Accuracy: 36.6 |
| common-sense-reasoning-on-commonsenseqa | Few-shot Direct GPT-J | #35 | Accuracy: 20.9 |