Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Language Modeling on SNLG and SNLU evaluation suites (test)

78.71SNLG Score

Teacher baseline (OPT-6.7B)

48.934856.664964.39572.1251Feb 19, 2024
Updated 4d ago

Evaluation Results

MethodLinks
2024.02
78.7123.4351.07
2024.02
74.2122.2848.25
2024.02
68.922.645.75
2024.02
62.6619.5641.11
2024.02
62.6219.2640.94
2024.02
62.4619.2140.84
2024.02
61.3619.0740.21
2024.02
61.319.5440.42
2024.02
61.2519.1840.22
2024.02
61.2417.4839.36
2024.02
61.2318.7740
2024.02
60.7719.1839.97
2024.02
60.7619.2940.02
2024.02
60.719.0239.86
2024.02
60.5916.8738.73
2024.02
60.0117.0838.55
2024.02
59.8817.4638.67
2024.02
59.8718.4139.14
2024.02
59.7419.4639.6
2024.02
59.0217.838.41
2024.02
58.8617.2838.07
2024.02
58.3320.3639.35
2024.02
57.5619.3138.43
2024.02
57.2619.4838.37
2024.02
56.7619.7338.25
2024.02
55.5717.9936.78
2024.02
55.4518.3336.89
2024.02
55.318.6937
2024.02
54.419.436.9
2024.02
54.3719.1636.76
2024.02
54.3419.2736.8
2024.02
52.6918.835.75
2024.02
52.3618.6635.51
2024.02
52.2718.3535.31
2024.02
52.1818.5735.37
2024.02
52.1619.5835.87
2024.02
51.918.5235.21
2024.02
51.8717.3234.59
2024.02
51.618.1534.87
2024.02
51.2618.5634.91
2024.02
50.8619.1334.99
2024.02
50.6218.8834.75
2024.02
50.5418.0534.3
2024.02
50.0818.3334.2