Run Megatron-LM on SLURM

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.

Source: github.com/NVIDIA/skills/tree/HEAD/skills/mcore-run-on-slurm · License: Other