Model
PaLM 540B
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| BoolQ | 88.7 | score | N/A | Independent | medium | arxiv.org |
| HellaSwag | 83.8 | accuracy | N/A | Independent | medium | arxiv.org |
| HellaSwag | 83.8 | overall_accuracy | N/A | Self-reported | medium | jmlr.org |
| HellaSwag | 83.6 | accuracy | N/A | Independent | medium | arxiv.org |
| HellaSwag | 83.4 | accuracy | N/A | Independent | medium | arxiv.org |
| LAMBADA | 89.7 | accuracy | N/A | Independent | medium | arxiv.org |
| LAMBADA | 81.8 | accuracy | N/A | Independent | medium | arxiv.org |
| LAMBADA | 77.9 | accuracy | N/A | Independent | medium | arxiv.org |
| LAMBADA | 77.9 | score | N/A | Independent | medium | arxiv.org |
| MMLU | 69.3 | exact_match | N/A | Independent | medium | arxiv.org |
| Natural Questions | 39.6 | exact_match | N/A | Independent | medium | arxiv.org |
| Natural Questions | 29.3 | exact_match | N/A | Independent | medium | arxiv.org |
| Natural Questions | 21.2 | exact_match | N/A | Independent | medium | arxiv.org |
| OpenBookQA | 68.0 | accuracy | N/A | Self-reported | medium | jmlr.org |
| OpenBookQA | 53.4 | accuracy | N/A | Independent | medium | arxiv.org |
| OpenBookQA | 53.4 | accuracy | N/A | Independent | medium | arxiv.org |
| PIQA | 82.3 | score | N/A | Independent | medium | arxiv.org |
| RTE | 79.6 | accuracy | N/A | Independent | medium | arxiv.org |
| RTE | 78.7 | accuracy | N/A | Independent | medium | arxiv.org |
| RTE | 72.9 | accuracy | N/A | Independent | medium | arxiv.org |
| StrategyQA | 76.4 | accuracy | N/A | Independent | medium | arxiv.org |
| Trivia QA | 81.4 | exact_match | N/A | Independent | medium | arxiv.org |
| Trivia QA | 81.4 | exact_match | N/A | Independent | medium | arxiv.org |
| Trivia QA | 81.4 | exact_match | N/A | Self-reported | medium | jmlr.org |
| Trivia QA | 76.9 | exact_match | N/A | Independent | medium | arxiv.org |
| WebQuestions | 43.5 | exact_match | N/A | Independent | medium | arxiv.org |
| WebQuestions | 22.6 | exact_match | N/A | Independent | medium | arxiv.org |
| WebQuestions | 10.6 | exact_match | N/A | Independent | medium | arxiv.org |
| WinoGrande | 85.1 | accuracy | N/A | Independent | medium | arxiv.org |
| WinoGrande | 85.1 | accuracy | N/A | Self-reported | medium | jmlr.org |
| WinoGrande | 81.1 | accuracy | N/A | Independent | medium | arxiv.org |
| WSC | 89.5 | accuracy | N/A | Independent | medium | arxiv.org |
| WSC | 89.1 | accuracy | N/A | Independent | medium | arxiv.org |
| WSC | 86.3 | accuracy | N/A | Independent | medium | arxiv.org |
Math
MathCoding
CodingMultilingual
Multilingual| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| MGSM | 55.0 | average_(%) | N/A | Independent | medium | arxiv.org |
| TyDiQA-GoldP | 52.9 | exact_match | N/A | Independent | medium | arxiv.org |
Reproduced by the community
Loading community runs…