Open Models Almanac
Methodstraining objectivelanguage modelling objective

single mention · training objective

Large-scale pre-training on 22 trillion tokens

Identifies pre-training by its 22-trillion-token training scale, without specifying a distinct objective.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See language modelling objective for the methods filed beside it.

Filed under training objective › language modelling objective