ASTERIZER/LUNA_PreTrain download history
ASTERIZER/LUNA_PreTrain is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 364 times (30 in the last 7 days), and 440 times in total. It ranks #42,322 among datasets by monthly downloads.
Dataset Card for LUNA Pretraining Corpus A deduplicated, English-only, tokenized pretraining corpus — ~4.5B tokens in 270 ready-to-train litdata chunks — built for the LUNA-100M causal language model. This repository is the pretraining data engine for the LUNA model family. It ships t