Metamath
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MetaMath Insufficient
72.5Success Rate (SR)
19
MetaMath 1k
212Token Count
14
Metamath (val)
56.5Performance
6
Metamath set.mm (val)
29.22Performance Score
3
Metamath (test)
65.6Pass@8
2