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

Common Sense Reasoning on ARC-C, ARC-E, BoolQ, HellaSwag, PIQA, WinoGrande (zero-shot)

77.67Average Accuracy (Zero-Shot)

FP16 Baseline

35.092446.146257.268.2538Feb 9, 2026Feb 16, 2026Feb 24, 2026Mar 3, 2026Mar 11, 2026Mar 18, 2026Mar 26, 2026
Updated 22d ago

Evaluation Results

MethodLinks
2026.03
77.6757.3480.4784.183.3282.6478.14
2026.03
77.1555.6378.0387.4384.0782.2675.45
2026.02
74.19------
2026.03
73.2449.5776.0584.7178.1379.7171.27
2026.02
73.19------
2026.02
72.68------
2026.02
72.64------
2026.02
72.64------
2026.02
72.62------
2026.02
71.84------
2026.02
71.68------
2026.02
71.66------
2026.02
71.54------
2026.02
71.251.7180.8981.7760.5678.8473.4
2026.02
71.251.7180.8981.7760.5678.8473.4
2026.02
71.1955.3883.586.7357.1276.567.88
2026.02
71.1955.3883.586.7357.1276.567.88
2026.02
70.6654.1882.6286.4555.776.3968.59
2026.02
70.5654.2782.7486.7356.6276.2266.77
2026.03
70.5446.4274.6278.0175.978.8469.46
2026.02
70.553.4182.3686.0256.3476.0668.82
2026.02
70.4453.7582.286.0256.3976.2268.03
2026.02
70.453.6782.3785.9355.9175.7968.75
2026.02
70.4------
2026.02
70.22------
2026.02
70.21------
2026.02
70.1649.1579.5981.1359.778.6772.69
2026.02
70.0648.779.2180.859.7378.6273.32
2026.02
69.9749.3279.1281.0459.7978.2472.3
2026.02
69.948.4678.5381.046078.1873.16
2026.02
69.7448.2978.8980.2159.4778.2973.32
2026.02
69.68------
2026.02
69.4------
2026.02
69.1546.2577.3681.9358.7677.5373.09
2026.02
68.6550.5680.9384.4352.875.967.25
2026.02
68.41------
2026.02
68.1650.3480.2685.1152.2975.365.67
2026.02
68.1650.3480.2685.1152.2975.365.67
2026.02
68.0847.2778.4184.3954.2376.2767.9
2026.02
68------
2026.02
67.6147.6977.6184.4653.7674.7667.35
2026.02
67.5347.5977.6284.0453.8874.5967.46
2026.02
67.47------
2026.02
67.46------
2026.02
67.22------
2026.02
66.8142.9275.4278.6955.7377.7570.32
2026.02
66.71------
2026.02
66.644.2874.6283.3353.8875.9567.56
2026.02
66.3343.8675.2176.1555.7975.9571.03
2026.02
65.9941.373.4876.4556.6777.3770.64
2026.02
65.9446.6578.2784.3149.8574.2162.35
2026.02
65.8741.8171.7176.7356.7176.8871.35
2026.02
65.8147.5777.982.4350.1274.4362.43
2026.02
65.746.2977.3283.2350.3774.162.88
2026.02
65.6441.3871.3477.6855.6776.1271.67
2026.02
65.54------
2026.03
65.2437.7164.1475.0273.0275.7965.75
2026.02
64.72------
2026.02
64.19------
2026.02
64.0944.2874.0382.9749.2872.861.17
2026.02
63.87------
2026.02
63.8------
2026.02
63.345.4875.4281.9944.7269.5362.67
2026.02
63.2------
2026.02
63.1941.8174.7983.4647.6173.1258.33
2026.03
62.7440.768.170.1462.1272.0963.3
2026.03
62.737.362.4572.1369.8473.2461.23
2026.02
62.0341.571.4180.9545.6172.6460.06
2026.02
61.7639.3370.9280.9546.8170.9261.64
2026.02
61.5134.2166.0471.8955.2673.5668.11
2026.02
61.1740.2768.2781.1946.0270.8760.38
2026.02
60.8139.2571.0980.7343.6470.2459.91
2026.03
60.7936.8663.4769.6359.4171.4463.93
2026.02
60.6640.2771.9377.3743.2271.3359.83
2026.03
60.3939.9361.2465.2665.4871.1659.27
2026.03
59.9337.259.5566.366.6371.7158.17
2026.02
59.42------
2026.03
59.332.9459.6467.7762.3371.3861.72
2026.02
56.99------
2026.02
56.3937.7965.3570.5843.3967.2553.99
2026.03
56.2633.4556.461.9660.4670.1855.09
2026.03
56.1533.6259.362.5455.0666.9259.43
2026.02
55.49------
2026.03
54.9531.9154.1763.6755.9568.4455.56
2026.02
53.8435.9558.6869.240.6567.7450.83
2026.03
49.9431.2346.2157.2852.6561.3750.91
2026.03
49.6931.0648.9953.6747.3162.5754.54
2026.03
49.0932.3448.1550.5549.9161.5952.01
2026.02
47.9124.1547.8167.8633.7262.7951.14
2026.03
45.2927.1339.7751.0741.3260.8851.54
2026.03
39.4824.433.8438.7833.7553.3252.8
2026.03
36.7326.0225.9738.8125.9250.4453.2
2026.03
-43.275.679.357.178.169.9
2026.03
-32.542.862.3-71.262.4
2026.03
-34.769.367.546.673.564.6
2026.03
-35.569.87349.775.266.9
2026.03
-19.42654.6--51.8
2026.03
-21.635.257.5--51.5
2026.03
-32.863.774.8--65.7
2026.03
-35.265.375.4--64.9
Showing 100 of 104 rows