// HACKER NEWS — CYBERSECURITY
Show HN: TurboGPT: train 22KiB transformer in 13s
Windows, Visual Studio 2022 C++ tools, and CUDA 13.4:
CudaArch is the GPU compute capability from NVIDIA's CUDA GPU list.
The run stores its checkpoint at
runs/ctx4/ctx4.pt, containing model, optimizer, scheduler, and trainer state.
Use --load CHECKPOINT.pt to resume it.
runs/ctx4/report.json is derived from the log directory. Logs are TensorBoard-compatible:
one report per batch, capped at 8Mi reports, and flushed with periodic or final checkpoints.