Gpt2
开源来自 OpenAI · 发布于 2022-03-02
6.5
平均分
N/A
输入价格
N/A
输出价格
N/A
上下文窗口
text-generation
类型
Tested on 6 benchmarks with 6.5% average. Top scores: IFEval (17.9%), MUSR (15.3%), BBH (HuggingFace) (2.7%).
基准测试分数
| 基准测试 | 类别 | 分数 | Bar |
|---|---|---|---|
| IFEval | language | 17.9 | |
| MUSR | reasoning | 15.3 | |
| BBH (HuggingFace) | general | 2.7 | |
| MMLU-PRO | knowledge | 1.8 | |
| GPQA | knowledge | 1.1 | |
| MATH Level 5 | math | 0.2 |