Show HN: TurboGPT: train 22KiB transformer in 13s

Automated excerpt

Tiny byte-level GPT training in CUDA C++. Windows, Visual Studio 2022 C++ tools, and CUDA 13. 4: CudaArch is the GPU compute capability from NVIDIA's CUDA GPU list. runs/ctx4/ctx4. pt, containing model, optimizer, scheduler, and trainer state. Logs are TensorBoard-compatible: one report per batch, capped at 8Mi reports, and flushed with periodic or final checkpoints. hn1g after 1. 5G training tokens: 2. 52435 BPB.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.