This is a GPT-2 based model that has been trained with Korean Wikipedia dataset.

Since there is no Korean pre-trained model that has been trained with a large dataset like Wikipedia for GPT-2 yet, so I made a decision to train GPT-2 for Korean texts.

It has been trained with Korean Wikipedia dataset (train wikipedia article count: 334420, validation wikipedia article count: 83605).

Yongwoo Jeong, Sep 13th, 2022.