GPT-2 (large)

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

Bias DetectiononStereoSet
ICAT Score· 2020-04-20
70.54
best: 72.97 (GPT-2 (small))
StereoSet: Measuring stereotypical bias in pretrained language models arXiv:2004.09456

Medical1 result

Language ModellingonWikiText-2
Test perplexity· uses extra data
19.93
best: 8.21 (SparseGPT (175B, 50% Sparsity))