Verified record · from the answer index
training-llms-megatron
skilltested · static-verifiedscore 49/100signed passport ✓909 uses
Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/c
Declared tools (1)
training-llms-megatron
skill:orchestra-research/ai-research-skills/training-llms-megatron · published by orchestra-research · source repository