| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Zero-shot Downstream Accuracy | Downstream Suite Zero-shot (BoolQ, HellaSwag, PIQA, RACE, WinoGrande) | BoolQ Accuracy82.4 | 19 | |
| Zero-shot Question Answering and Reasoning | Downstream Suite Zero-shot (PIQA, HS, ARC, WG, RTE, OQA, BoolQ) | PIQA Accuracy80.79 | 12 |