Common Sense Reasoning Tasks
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Common Sense Reasoning Tasks
93Avg Score
321
Common Sense Reasoning Tasks (ARC-C, ARC-E, BoolQ, HellaSwag, PIQA, WinoGrande) zero-shot
77.67Average Accuracy (Zero-Shot)
92
9 Common Sense Reasoning Tasks (WinoGrande, SocialIQA, LAMBADA, MMLU, ARC-Easy, ARC-Challenge, HellaSwag, OpenBookQA, PIQA) Average
72.7Accuracy
65
5 common-sense reasoning tasks Llama-3-8B
87.07Average Accuracy
35
Common-Sense Reasoning Tasks
16Wiki PPL
18
Common-sense reasoning tasks (ARC-C, ARC-E, HellaSwag, Lambada, PIQA, WinoGrande) (test)
44.88ARC-C Accuracy
16
5 common-sense reasoning tasks Llama-2-70B
72.41Average Accuracy
15
5 common-sense reasoning tasks Llama-2-13B
67.81Accuracy
15
5 common-sense reasoning tasks Llama-3-70B
75.33Average Accuracy
9
Common-sense reasoning tasks WikiText, LAMBADA, PIQA, HellaSwag, WinoGrande, ARC
25.89PPL (WikiText)
8