mishig/autoresearch-optimizer-findings download history
mishig/autoresearch-optimizer-findings is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 238 times (35 in the last 7 days), and 838 times in total. It ranks #58,794 among datasets by monthly downloads.
Ablating Optimizer Ideas Around MuonAdamW for Transformer Pretraining Overview 27 optimizer experiments on a 50M-parameter transformer, systematically testing whether anything can beat a well-tuned MuonAdamW baseline. We tried curvature-aware momentum, spectral boosting via pow