HyperAI超神经

Multiple Choice Question Answering Mcqa On 15

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

模型名称
Accuracy
Paper TitleRepository
Chinchilla (few-shot, k=5)51.0Galactica: A Large Language Model for Science
GAL 120B (zero-shot)49Galactica: A Large Language Model for Science
BLOOM (few-shot, k=5)6.0Galactica: A Large Language Model for Science
OPT (few-shot, k=5)17.0Galactica: A Large Language Model for Science
0 of 4 row(s) selected.