v8.3.13: Bug fixes and validation improvements

Latest

Latest

github-actions released this 23 Mar 07:20

· 5 commits to v8.3.x since this release

f72185a

Add clear error message for float64 dtype mismatch in BLIS gemm (#461).
Document that Model.from_disk requires matching architecture (#751).
Document that Loss.get_grad computes gradient w.r.t. logits, not
post-softmax probabilities (#901).
Remove stale comments about torch.cuda.amp.autocast migration (#967).
Fix numpy deprecation warning filter syntax in pyproject.toml.
Ignore NumPy 2.4 align deprecation from pickled dtype in test data.

Assets 42