Commit Graph

2 Commits

Author SHA1 Message Date
Wing Lian
c0a0c7534c Activation checkpointing with offloading to disk with prefetch (#2663)
* offload activations to disk instead of CPU RAM

* add prefetch

* Disco :dance:

* include offload_disk in e2e test for AC

* document and make sure to cleanup

* fix annotation to match docs

* fix docs build

* address PR feedback
2025-05-13 16:39:39 -04:00
Wing Lian
caf5cb63ea add e2e smoke test for using activation/gradient checkpointing with offload (#2565)
* add e2e smoke test for using activation/gradient checkpointing with offload

* disable duplicate code check for the test

* fix relative import

* seq len too small to test this dataset with packing

* Fix checkpoint ptaching for tests
2025-04-25 21:11:17 -04:00