Registry
SKILL
mcore-run-on-slurm
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
- Identifier
- urn:air:outshift.io:agntcy:mcore-run-on-slurm
- Catalog host
- outshift.io · via agntcy registry
- Anchor check
- Not anchored
- Last crawled
- seen 1h ago
Discovered through outshift.io's registry, not anchored by StealthStack. We relay the listing as-is; we have not checked that the URN authority matches the publishing host. How trust works →