This model is a pretrained revision of KoGPT6B-float16 on causal language modeling (CLM) objective, which was fitted using unlabeled context from [a small supervised tuning dataset for summarization](https://huggingface.co/datasets/sysong11/sum_train_rev).