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

Question Answering on MMLU (Vanilla and Absolute Accuracy)

65.98Vanilla Accuracy

GPT-3.5

21.551233.085644.6256.1544Mar 14, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.03
65.9852.09
2024.03
62.3546.76
2024.03
61.8647.31
2024.03
58.7344.39
2024.03
52.9837.24
2024.03
51.8735.72
2024.03
51.6835.21
2024.03
50.1133.68
2024.03
49.432.5
2024.03
49.3432.21
2024.03
49.3429.47
2024.03
47.2731.11
2024.03
41.6525.48
2024.03
23.2611.55