mcore-run-on-slurm

Skill

outshift.io · via agntcy registry Unverified — relayed by outshift.io seen 5h ago

About

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.

Capabilities

The crawler did not record capability metadata for this resource. Inspect the endpoint directly to see what it exposes.

Provenance

Discovered Relayed by agntcy

URN authority urn:air:outshift.io:agntcy:mcore-run-on-slurm

Catalog host outshift.io

Anchor check Not anchored

Last crawled seen 5h ago

mcore-run-on-slurm

About

Capabilities

Provenance

Tags