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

ARC, GSM8k, HellaSwag, MMLU

Benchmarks

Task NameDataset NameSOTA ResultTrend
Instruction Tuning EvaluationARC, GSM8k, HellaSwag, MMLU (test val)
ARC Accuracy52.31
7
Showing 1 of 1 rows