single mention · data curation
Log-linear proxy model for predicting benchmark BPB from data mixture
Fitting a log-linear law to proxy-model results to predict benchmark BPB for candidate mixtures.
The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See data mixture & curriculum for the methods filed beside it.
Also called log-linear law.
Filed under data curation › data mixture & curriculum