Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

WebText

Benchmarks

Task NameDataset NameSOTA ResultTrend
Language ModelingWebText (test)
Diversity (Div)0.87
14
Text GenerationWebText
Perplexity (PPL)0.05
9
Language GenerationWebText (completions)
Perplexity (PPL)10.16
7
Text Generation Evaluation CorrelationWebText (test)
Perplexity (PPL)0.643
3
Open-ended Text GenerationWebText (test)
Same Preference Count97
2
Showing 5 of 5 rows