CausalNLP/gpt2-training-ar-zh-ko-ja-4b download history

CausalNLP/gpt2-training-ar-zh-ko-ja-4b is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 197 times (86 in the last 7 days), and 470 times in total. It ranks #68,386 among datasets by monthly downloads.

Balanced Arabic-Chinese-Korean-Japanese 4B-token training data Sequential FineWeb-2 documents tokenized with CausalNLP/gpt2-tokenizer-ar-zh-ko-ja. Each language contains at least 1,000,000,000 tokens in complete documents. Total target: 4,000,000,000 tokens. Approximately 100,000,000

Open CausalNLP/gpt2-training-ar-zh-ko-ja-4b on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.