TranslatedMNBVC (Massive Never-ending BT Vast Chinese corpus), an ultra-large-scale Chinese corpus. It benchmarks against the 40T of data used to train ChatGPT. The MNBVC dataset includes not only mainstream culture, but also data from various niche cultures and even "Martian language" (internet slang). The MNBVC dataset includes plain-text Chinese data in every form, such as news, essays, novels, books, magazines, papers, lines from scripts, forum posts, wikis, classical poems, song lyrics, product descriptions, jokes, embarrassing stories, and chat logs.
- chinese
- chinese-language
- chinese-nlp
- chinese-simplified