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

Zero-shot Reasoning on (OpenbookQA, ARC-e, ARC-c, WinoGrande, HellaSwag, PIQA, MathQA)

69.99Average Accuracy

Meta-Llama-3-8B Dense

-1.759616.867735.49554.1223Feb 3, 2026Feb 7, 2026Feb 12, 2026Feb 16, 2026Feb 21, 2026Feb 25, 2026Mar 2, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
69.994577.6953.3372.8579.1780.79--69.6881.44
2026.03
6740.873.8648.2971.1974.9278.84--68.5979.51
2026.03
66.2740.874.1245.6570.4874.2978.78--66.7979.24
2026.03
66.2341.673.2348.7271.974.479.33--61.7378.9
2026.03
65.6843.47145.5672.4574.8578.51--61.7377.92
2026.03
64.9941.470.4543.8671.3573.4578.18--62.4578.81
2026.03
64.9441.269.1943.5271.9872.977.37--67.1576.24
2026.03
64.3239.867.0942.9270.5672.477.48--67.1577.13
2026.03
64.2240.470.9245.3970.7272.3177.8--60.6575.6
2026.03
64.1240.467.843.670.0172.5779--63.975.69
2026.03
63.9140.271.0444.4572.7771.1976.82--56.6878.13
2026.03
63.8140.669.9944.2869.7771.2577.69--58.8478.07
2026.03
63.6740.467.3841.8970.8872.5178.24--62.4575.63
2026.03
63.6439.268.1441.8971.9873.1778.73--59.5776.42
2026.03
63.2539.468.8641.7270.0171.6277.04--61.3776.02
2026.03
62.5438.865.6640.768.2769.5477.15--66.4373.73
2026.03
61.4235.262.0437.5467.2566.575.68--70.476.79
2026.03
61.3637.663.1739.0868.0368.1575.41--62.0977.34
2026.03
60.5237.260.3536.4368.6766.8674.59--65.3474.71
2026.03
59.9936.261.6637.0368.9868.3475.9--62.4569.33
2026.03
59.8738.263.5538.6567.4867.4774.7--55.673.27
2026.03
59.236.860.6537.4666.5465.4373.61--57.475.72
2026.03
58.3355837.266.6964.7773.94--61.0169.79
2026.03
56.153557.4535.5863.9361.272.36--53.0770.58
2026.02
54.92----------
2026.02
54.24----------
2026.02
52.3----------
2026.02
52286738675678270--
2026.03
51.3332.245.9226.9660.6952.9768.12--53.4370.34
2026.02
51.01----------
2026.02
50.63----------
2026.03
50.231.246.8427.9960.5452.1867.95--53.4361.44
2026.03
50.1530.644.0226.3760.0649.6366.81--55.668.1
2026.02
5029683665457525390--
2026.03
48.9629.843.7726.3758.2548.4966--52.7166.27
2026.02
47.32----------
2026.03
47.1130.642.0925.2652.6443.8165.61--52.7164.13
2026.02
4727633364457124920--
2026.03
46.8529.643.7326.6259.4348.4266.38--53.4347.16
2026.03
46.5829.440.2824.9154.8543.362.35--53.0764.46
2026.02
46265931664470231,080--
2026.03
44.4927.238.7224.2351.9336.8862.24--52.7162.02
2026.02
44225829634369241,470--
2026.02
43255327644168241,670--
2026.02
41235029623665232,110--
2026.02
40225030613664222,280--
2026.02
38224127583461232,630--
2026.02
37194225583360212,830--
2026.02
34163325523054233,410--
2026.02
32153120522854223,800--
2026.02
31142822502755213,990--
2026.02
30132822482655194,180--
2026.02
29122621492653184,380--
2026.02
29116581059,600--
2026.02
265220539,660--
2026.02
161001109,780--