Kermt Pretrain Scratch

Pretrain a fresh KERMT model from scratch on a user-provided corpus. Builds a new vocabulary from the corpus, instantiates the model architecture from defaults, and launches pretrain_ddp.py inside the kermt container (detached for long runs). Unlike kermt-continue-pretrain, no starting checkpoint is loaded — the model is randomly initialized.

NVIDIA Updated 23k repo stars

File contents

NVIDIA/skills commit c1483363ee

Frequently asked questions

npx get-skill add nvidia/kermt-pretrain-scratch