Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Common Crawl

Benchmarks

Task NameDataset NameSOTA ResultTrend
Language ModelingCommon Crawl CC-10B (held-out)
CC Loss2.9421
4
Language IdentificationCommon Crawl (CC) (test)
Macro Precision72.59
3
Text SimilarityCC (Common Crawl)
BLEU89
2
Showing 3 of 3 rows