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

Classification on Zero-shot Evaluation Suite (BoolQ, PIQA, HellaSwag, WinoGrande, ARC-e, ARC-c, OBQA)

69.31Average Accuracy (Zero-Shot Suite)

Dense

39.264447.064754.86562.6653May 28, 2023Nov 9, 2023Apr 23, 2024Oct 6, 2024Mar 21, 2025Sep 3, 2025Feb 16, 2026
Updated 2d ago

Evaluation Results

MethodLinks
2026.02
69.31-------
2026.02
66.85-------
2026.02
66.79-------
2026.02
64.39-------
2026.02
64.34-------
2026.02
64.14-------
2026.02
63.52-------
2023.05
63.2573.1878.3572.9967.0167.4541.3842.4
2026.02
63.21-------
2026.02
63.21-------
2026.02
62.79-------
2026.02
62.54-------
2026.02
62.54-------
2026.02
62.32-------
2026.02
60.87-------
2026.02
60.6-------
2023.05
60.0565.6279.317062.7665.8737.6939.14
2026.02
59.82-------
2026.02
59.62-------
2023.05
59.4665.3776.6569.4163.7865.4536.1239.5
2026.02
59.33-------
2023.05
59.2364.6277.268.863.1464.3136.7739.8
2026.02
58.97-------
2026.02
58.59-------
2026.02
58.31-------
2026.02
58.06-------
2026.02
57.8-------
2026.02
57.79-------
2023.05
57.2365.7574.764.5259.3560.6536.2639.4
2026.02
56.08-------
2026.02
55.96-------
2026.02
55.88-------
2026.02
54.88-------
2026.02
54.26-------
2026.02
54.19-------
2023.05
53.5961.8970.8158.3456.8754.8734.0238.4
2026.02
53.44-------
2026.02
52.98-------
2026.02
52.56-------
2026.02
50.39-------
2026.02
50.11-------
2023.05
49.7161.8871.5347.8655.0145.1331.6234.98
2023.05
49.5661.4370.8847.6555.1245.7830.535.62
2026.02
49.34-------
2023.05
48.6960.2869.3147.0653.4345.9629.1835.6
2026.02
47.93-------
2026.02
47.88-------
2026.02
47.45-------
2026.02
47.19-------
2026.02
46.77-------
2026.02
46.74-------
2026.02
46.72-------
2026.02
46.26-------
2026.02
46.14-------
2023.05
45.4350.957.3838.1255.9842.6834.238.78
2026.02
44.87-------
2026.02
42.46-------
2026.02
42.4-------
2023.05
40.4247.454.3633.4953.137.8826.630.12