Skip to content

Releases: JonasGeiping/cramming

New Torch 2.1 Version

13 Jun 16:46
Compare
Choose a tag to compare

This release is the new version for torch 2.1. The code is nicer to read, has fewer dependencies (no more flash attention installations), data can now be easily streamed, and training is faster.

The new checkpoints are about 2% better on GLUE with the same budget.

Old Version

13 Jun 16:44
4a5e300
Compare
Choose a tag to compare
Old Version Pre-release
Pre-release

This release is the old version, usable with PyTorch 1.13.