mrq/bitsandbytes-rocm
1
0
Fork 1
Code Issues 1 Pull Requests Packages Projects Releases Wiki Activity

Added changes for analysis Adam.

Browse Source
This commit is contained in:
Tim Dettmers 2021-10-21 16:07:24 -07:00
parent 0c5fa5a637
commit d2f1672690
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
CHANGELOG.md
View File

@ -28,6 +28,7 @@ Features:
Features: Features:
- Added `skip_zeros` for block-wise and 32-bit optimizers. This ensures correct updates for sparse gradients and sparse models. - Added `skip_zeros` for block-wise and 32-bit optimizers. This ensures correct updates for sparse gradients and sparse models.
- Added support for Kepler GPUs. (#4) - Added support for Kepler GPUs. (#4)
- Added Analysis Adam to track 8-bit vs 32-bit quantization errors over time.
Bug fixes: Bug fixes:
- fixed "undefined symbol: \_\_fatbinwrap_38" error for P100 GPUs on CUDA 10.1 (#5) - fixed "undefined symbol: \_\_fatbinwrap_38" error for P100 GPUs on CUDA 10.1 (#5)