BEE-spoke-data/UltraTextbooks-2.1-fw_mix download history

BEE-spoke-data/UltraTextbooks-2.1-fw_mix is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 743 times (109 in the last 7 days), and 8,098 times in total. It ranks #24,254 among datasets by monthly downloads.

BEE-spoke-data/UltraTextbooks-2.1-fw_mix filtered ultratextbooks for min 50 words shuffle in 500k rows from fineweb to facilitate continual pretrain GPT-4 tiktoken token count: token_count count 3.701646e+06 mean 9.934539e+02 std 1.726200e+03 min 5.400000e+01 25% 2.5

Models trained on UltraTextbooks-2.1-fw_mix

3 models list it as training data.

• BEE-spoke-data/mega-ar-350m-L3t-v0.08-ultraTBfw 25 downloads in 30 days
• pszemraj/mega-ar-525m-v0.07-ultraTBfw 22 downloads in 30 days
• pszemraj/mega-ar-350m-v0.13 21 downloads in 30 days

Open BEE-spoke-data/UltraTextbooks-2.1-fw_mix on Hugging Face

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