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

Greater-Than task

Benchmarks

Task NameDataset NameSOTA ResultTrend
Head RecoveryGreater-Than task Pythia-410M (N=200)
Hit Rate @580.4
5
Head RecoveryGreater-Than task GPT-2 (N=200)
Accuracy @583.2
5
Showing 2 of 2 rows