Repository

Megatron-LM

Verified repository

Ongoing research training transformer models at scale

NVIDIA/Megatron-LM17Kstars4.2Kforks

13

skills

Jul 17

updated

Published by

NVIDIAVerified publisherwww.nvidia.com

Skills from this repository

Search the skills in this repository.

Sort by

Showing 13 skills from NVIDIA/Megatron-LM

SkillTagsUpdated

mcore-build-and-dependency
manage Megatron-LM development environments
ContainersDeploymentPython
Jul 14
mcore-bump-base-image
update NVIDIA PyTorch base images
CI/CDDeployment
Jul 14
mcore-cicd
manage CI/CD pipelines for Megatron-LM
CI/CDDeploymentGitHub
Jul 14
mcore-create-issue
investigate CI failures and create issues
DebuggingGitHubTriage
Jul 14
mcore-linting-and-formatting
lint and format Megatron-LM code
Best PracticesCode Analysis
Jul 14
mcore-migrate-gpt-to-hybrid
migrate Megatron-LM models to HybridModel
Machine LearningMigrationNVIDIA
Jul 17
mcore-onboard-gb200-1node-tests
onboard functional tests for GB200
QATesting
Jul 14
mcore-run-on-slurm
launch distributed training jobs on SLURM
DeploymentInfrastructure
Jul 14
mcore-split-pr
split pull requests to reduce review load
Code ReviewGitHubPull Requests
Jul 14
mcore-testing
run and manage Megatron-LM tests
QATesting
Jul 14
nightly-sync
manage nightly main-to-dev sync workflows
AutomationCI/CD
Jul 14
respond-to-issue
draft responses to GitHub issues
CommunicationsGitHubWriting
Jul 14
update-golden-values
update and score golden test values
GitHub ActionsQATesting
Jul 14