Common Sense Reasoning On Big Bench Known
评估指标
Accuracy
评测结果
各个模型在此基准测试上的表现结果
| Paper Title | Repository | ||
|---|---|---|---|
| PaLM-540B (few-shot, k=5) | 73.9 | PaLM: Scaling Language Modeling with Pathways | |
| Chinchilla-70B (few-shot, k=5) | 65.2 | Training Compute-Optimal Large Language Models | |
| Gopher-280B (few-shot, k=5) | 63.6 | Scaling Language Models: Methods, Analysis & Insights from Training Gopher |
0 of 3 row(s) selected.