Model
Megatron-Turing NLG 530B
Microsoft,Nvidia
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| ANLI | 39.7 | score | N/A | Self-reported | medium | arxiv.org |
| BoolQ | 84.8 | score | N/A | Self-reported | medium | arxiv.org |
| BoolQ | 78.2 | score | N/A | Independent | medium | arxiv.org |
| HellaSwag | 82.4 | overall_accuracy | N/A | Independent | medium | arxiv.org |
| HellaSwag | 82.4 | overall_accuracy | N/A | Self-reported | medium | arxiv.org |
| LAMBADA | 87.2 | score | N/A | Self-reported | medium | arxiv.org |
| LAMBADA | 76.6 | score | N/A | Independent | medium | arxiv.org |
| PIQA | 83.2 | score | N/A | Self-reported | medium | arxiv.org |
| PIQA | 82.0 | score | N/A | Independent | medium | arxiv.org |
| WinoGrande | 78.9 | accuracy | N/A | Independent | medium | arxiv.org |
| WinoGrande | 73.7 | accuracy | N/A | Self-reported | medium | arxiv.org |
Reproduced by the community
Loading community runs…