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

Common-sense Reasoning on 5 common-sense reasoning tasks Llama-3-8B

87.07Average Accuracy

FULL-SFT

18.565236.350154.13571.9199Jul 10, 2024Nov 7, 2024Mar 7, 2025Jul 5, 2025Nov 2, 2025Mar 2, 2026Jul 1, 2026
Updated 23d ago

Evaluation Results

MethodLinks
2025.12
87.0782.9492.6874.9896.7189.490.2683.2186.11
2025.12
86.3480.9791.6275.5695.988.489.1182.5986.58
2025.12
86.1681.1491.3375.4495.7188.488.6881.5387.06
2025.12
86.1481.4891.5876.1795.6686.88981.4286.97
2025.12
86.1381.6591.0776.695.78788.5781.2687.21
2025.12
86.0880.291.0775.9996.0587.689.3381.1687.21
2025.12
86.0381.6691.4674.895.6687.289.0681.9986.42
2025.12
85.8980.7291.4175.0895.8786.489.1781.0687.37
2025.12
85.7579.7891.2575.696.0386.888.5281.3786.66
2025.12
85.7481.2391.4175.5795.728688.0881.9385.95
2025.12
85.7280.3890.9174.9596.1187.488.5280.0487.45
2025.12
85.6980.2990.8275.9695.7487.289.0680.685.87
2025.12
85.6379.6991.2575.9395.886.488.7480.4586.74
2025.12
85.6180.3791.0775.1995.2385.689.1181.1187.21
2025.12
85.6179.5290.7475.3595.7585.489.8881.2187.05
2025.12
85.5780.0390.9974.5696.038788.6381.1786.11
2025.12
85.4580.4690.4575.6695.778587.9281.0687.29
2026.07
70.868.785.964.776.465.679.667.358.4
2024.07
68.58--------
2026.07
68.26583.364.570.760.879.963.557.7
2024.07
67.35--------
2026.07
65.760.580.362.269.960.27761.853.5
2024.07
64.82--------
2026.07
64.160.379.663.665.658.273.158.653.5
2024.07
64.09--------
2026.07
63.960.278.362.56853.676.759.252.6
2026.07
63.161.4786363.650.876.757.353.6
2024.07
60.76--------
2024.07
60.58--------
2024.07
59.37--------
2024.07
58.72--------
2026.07
57.651.969.662.648.35370.254.550.6
2024.07
52.66--------
2024.07
51.74--------
2026.07
21.21012.561.912.323.447.42.20