Model
Phi-2
Microsoft · open weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| ANLI | 42.5 | score | N/A | Self-reported | medium | arxiv.org |
| BBH | 59.4 | average | N/A | Self-reported | medium | arxiv.org |
| HellaSwag | 53.6 | overall_accuracy | N/A | Self-reported | medium | arxiv.org |
| MMLU | 58.4 | exact_match | N/A | Independent | medium | crfm.stanford.edu |
| MMLU | 56.3 | exact_match | N/A | Self-reported | medium | arxiv.org |
| OpenBookQA | 73.6 | accuracy | N/A | Self-reported | medium | arxiv.org |
| Trivia QA | 45.2 | exact_match | N/A | Self-reported | medium | arxiv.org |
| WinoGrande | 54.7 | accuracy | N/A | Self-reported | medium | arxiv.org |
Coding
CodingReproduced by the community
Loading community runs…