HyperAIHyperAI

Command Palette

Search for a command to run...

各模型在此基准测试上的性能结果

指标

CW
MMLU
RoPE
Simi
Tokens
Base Model
MMLU Ratio
PE
BBH
Data
MATH
MBPP
GSM8K
Method
LAMBADA
HUMANEVAL
12 总计