Claude 3 Sonnet (5-shot)

Reported on 3 benchmarks across 3 tasks

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology2 results

Transfer LearningonMML
Average (%)
79
best: 87 (GPT-4 o1(300b))
Multi-Task LearningonMML
Average (%)
79
best: 87 (GPT-4 o1(300b))

Natural Language Processing1 result

Common Sense ReasoningonWinoGrande
Accuracy
75.1
best: 96.1 (ST-MoE-32B 269B (fine-tuned))