AverageMetaheuristicsEnjoyer/mobe-factor-cache-gpt-oss-20b-b8 download history

AverageMetaheuristicsEnjoyer/mobe-factor-cache-gpt-oss-20b-b8 is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 14 times (3 in the last 7 days), and 136 times in total. It ranks #470,174 among datasets by monthly downloads.

MoBE factor cache — GPT-OSS-20B (num_B=8) Trained MoBE (Mixture-of-Basis-Experts) factors for the MoE gate_proj / up_proj matrices of GPT-OSS-20B, plus the perplexity / activation-geometry results of the compression sweep that produced them. Contents mobe_factor_cache/

Open AverageMetaheuristicsEnjoyer/mobe-factor-cache-gpt-oss-20b-b8 on Hugging Face

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