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

Downstream Tasks Suite

Benchmarks

Task NameDataset NameSOTA ResultTrend
Downstream Task EvaluationDownstream Tasks Suite (LAMBADA, HellaSwag, PIQA, WinoGrande, OpenBookQA, ARC) (test)
HellaSwag Accuracy52.21
3
Showing 1 of 1 rows