| Benchmark | Model | Rank | Results |
|---|---|---|---|
| language-modelling-on-lambada | RWKV-4 14B | #20 | Accuracy: 71.05Perplexity: 3.81 |
| natural-language-inference-on-wnli | RWKV-4-Raven-14B | #22 | Accuracy: 49.3 |
| question-answering-on-piqa | RWKV-4 14B | #85 | Accuracy: 77.42 |
| common-sense-reasoning-on-winogrande | RWKV-4 14B | #103 | Accuracy: 62.98 |
| sentence-completion-on-hellaswag | RWKV-4 14B | #110 | Accuracy: 70.24 |