HyperAI
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
Correct
Dataset
Base LLMs
Irrelevant
Simplified CoT
Instructed LLMs
Detailed CoT (GPT4o)
Nuclear Norm
Gradient Norm
Spectral Radius
Math
STEM
Coding
Overall
Writing
Roleplay
Reasoning
Extraction
Humanities
3 行 总计