| Age | Commit message (Expand) | Author |
|---|---|---|
| 30 hours | Add MLX_EAGER_EVAL flag to further reduce memory pressure by force-evaluating... | sandrone |
| 46 hours | Update train_gpt_mlx.py | Will DePue |
| 47 hours | Merge pull request #32 from yhn112/fix-mlx-eval-memory-growth | Will DePue |
| 47 hours | Log MLX validation progress | Michael Diskin |
| 48 hours | Fix MLX validation loss accumulation | Michael Diskin |
| 2 days | match timing to main script to exclude eval timing | bernhardwalser |
| 2 days | Launch snapshot | Will DePue |
