finetune gpt-2

Train 774M, 1.5B models with the Google's S3 optimizer

LinksGitHub
Published
Views37 from Taiyuan, Ashburn, Miami

Fine-tuning scripts for GPT-2 774M and 1.5B parameter models using Google's S3 optimizer. One of the early open-source efforts to make large language model fine-tuning accessible.

Comments