raihan-js/vocab-tax-grid download history

raihan-js/vocab-tax-grid is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 22 times (22 in the last 7 days), and 22 times in total. It ranks #349,066 among datasets by monthly downloads.

Vocab Tax Grid Compute-matched vocabulary-size study for small code models. Fertility table: tokens/byte for 2k/8k/16k/32k BPE on TS/JS eval.json: bits-per-byte for 16 runs (3 sizes x 4 vocabs + 4 second seeds) grid.json: training logs (steps, loss, FLOPs) Key finding: 8k-16k wins a

Open raihan-js/vocab-tax-grid on Hugging Face

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