Open Models Almanac
Methodsoptimizationtraining parallelism

single mention · optimization

ZeRO-2 with strided activation checkpointing

This training setup combines ZeRO-2 sharding with strided activation checkpointing.

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

Filed under optimization › training parallelism