Gopher 280B (0-shot)

Reported on 3 benchmarks across 3 tasks · 1 paper

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

Natural Language Processing3 results

Question AnsweringonPIQA
Accuracy· 2021-12-08
81.8
best: 90.1 (Unicorn 11B (fine-tuned))
Scaling Language Models: Methods, Analysis & Insights from Training Gopher arXiv:2112.11446
Common Sense ReasoningonWinoGrande
Accuracy· 2021-12-08
70.1
best: 96.1 (ST-MoE-32B 269B (fine-tuned))
Scaling Language Models: Methods, Analysis & Insights from Training Gopher arXiv:2112.11446
Sentence CompletiononHellaSwag
Accuracy· 2021-12-08
79.2
best: 96.1 (CompassMTL 567M with Tailor)
Scaling Language Models: Methods, Analysis & Insights from Training Gopher arXiv:2112.11446