GLaM 62B/64E (One-Shot)

Reported on 2 benchmarks across 2 tasks · 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

Question AnsweringonNatural Questions
EM· 2021-12-13
26.3
best: 64 (Atlas (full, Wiki-dec-2018 index))
GLaM: Efficient Scaling of Language Models with Mixture-of-Experts arXiv:2112.06905

Medical1 result

Language ModellingonLAMBADA
Accuracy· 2021-12-13
80.9
best: 89.7 (PaLM-540B (Few-Shot))
GLaM: Efficient Scaling of Language Models with Mixture-of-Experts arXiv:2112.06905