GPT-3 175B (one-shot)

Reported on 1 benchmark across 1 task · 1 paper

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

Natural Language Processing1 result

Common Sense ReasoningonReCoRD
F1· 2022-05-24
90.2
best: 96.4 (Turing NLR v5 XXL 5.4B (fine-tuned))
Large Language Models are Zero-Shot Reasoners arXiv:2205.11916