HyperAI超神经

Auto Debugging On Big Bench Lite

评估指标

Exact string match

评测结果

各个模型在此基准测试上的表现结果

模型名称
Exact string match
Paper TitleRepository
PaLM 62B (few-shot, k=5)38.2PaLM: Scaling Language Modeling with Pathways
PaLM 8B (few-shot, k=5)14.7PaLM: Scaling Language Modeling with Pathways
PaLM 540B (few-shot, k=5)38.2PaLM: Scaling Language Modeling with Pathways
0 of 3 row(s) selected.