ResearchTasksMultiple Choice EvaluationMultiple Choice EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedDownstream Tasks (ARC, HellaSwag, PIQA, Winogrande) Standard LLM Eval (val/test)L3 Transformer56.98Average Accuracy9Feb 26, 2026
Downstream Tasks (ARC, HellaSwag, PIQA, Winogrande) Standard LLM Eval (val/test)L3 Transformer56.98Average Accuracy9Feb 26, 2026