Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Mathematical Reasoning on GSM8K @MetaMath (test)

72.3Accuracy

BF16

3.930421.680239.4357.1798Feb 16, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.02
72.3
2024.02
69.69
2024.02
68.75
2024.02
68.67
2024.02
68.51
2024.02
68.28
2024.02
66.62
2024.02
66.41
2024.02
64.38
2024.02
62.34
2024.02
62.11
2024.02
61.7
2024.02
61.33
2024.02
60.78
2024.02
59.3
2024.02
51.02
2024.02
36.64
2024.02
36.4
2024.02
32.5
2024.02
26.25
2024.02
23.67
2024.02
23.13
2024.02
21.87
2024.02
17.5
2024.02
16.09
2024.02
11.43
2024.02
9.45
2024.02
7.89
2024.02
7.89
2024.02
6.72
2024.02
6.56