mcore-run-on-slurm
DevOps1 installation(s)How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.
Ce skill s'installe en un clic sur un agent AIbergement (runtime OpenClaw) — il sera chargé et utilisé automatiquement lors des prochains tours de votre agent.