Commit message (Expand) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Improved launch config for layer-norm/rms-norm. (#2591) | Laurent Mazare | 2024-11-04 | 1 | -0/+45 |
* | Add the layernorm specialized op. (#2212) | Laurent Mazare | 2024-05-24 | 1 | -0/+27 |
* | Fix sigmoid gradient calculation and move sigmoid into a specialized op (#2114) | MilkFather | 2024-04-29 | 1 | -0/+11 |
* | Add the rope THD kernel. (#2014) | Laurent Mazare | 2024-04-05 | 1 | -0/+31 |
* | Contiguous variant of the rope kernel. (#1929) | Laurent Mazare | 2024-03-25 | 1 | -2/+30 |
* | Fast kernels for rotary embeddings. (#1928) | Laurent Mazare | 2024-03-24 | 1 | -0/+28 |
* | Custom op for RmsNorm (#1890) | Laurent Mazare | 2024-03-21 | 1 | -4/+30 |
* | Add a custom softmax implementation. (#744) | Laurent Mazare | 2023-09-05 | 1 | -0/+10 |
* | Move the test-utils bits to a shared place. (#619) | Laurent Mazare | 2023-08-27 | 1 | -7/+4 |
* | Add a yolo-v3 example. (#528) | Laurent Mazare | 2023-08-20 | 1 | -0/+3 |
* | Add the AdamW optimizer. (#307) | Laurent Mazare | 2023-08-02 | 1 | -14/+6 |
* | Softmax numerical stability. (#267) | Laurent Mazare | 2023-07-28 | 1 | -0/+62 |