General Understanding Tasks
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
General Understanding Tasks ARC-E, BoolQ, Wino., PIQA, HellaSwag, TruthfulQA, OBQA, LogiQA
64.1ARC-E Accuracy
8
General Understanding Tasks Chinese (ZH) Translated
45.92Accuracy
3
General Understanding Tasks Japanese (JA) Translated
44Average Accuracy
3
General Understanding Tasks French (FR) Translated
47.52Avg Accuracy
3
General Understanding Tasks German (DE) Translated
47.16Accuracy
3
General Understanding Tasks English (EN) Translated
61.07Avg Accuracy
2