Each module is a self-contained Kaggle notebook. Work through them in order — each one builds on the last. modules/ ├── 00_setup_and_profiling/ ← Start here. Verify your environment. ├── ...
To understand why TSP is important, you must first understand the two parallelism strategies it folds together. Tensor Parallelism (TP) splits model weights across GPUs. If you have a weight matrix in ...
DeepSpeed offers a collection of system technologies, that has made it possible to train models at these scales. The best technology to train your large model depends on various factors such as the ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results