これはひらがなに変換して逆向きに並べ替えたデータセットで事前学習した言語モデルです。 ひらがなを文字単位でトークンに分割しているため、回文や川柳のような音の数を重視するタスクに適しています。

This is a language model pre-trained on a dataset converted into Japaneses-Hiragana and reversed. Since it tokenizes Hiragana at the character level, it is suitable for tasks that emphasize the number of sounds, such as palindromes or senryu (a form of Japanese poetry).

Downloads last month
303
Safetensors
Model size
21.1M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.

Dataset used to train hukuda222/hiragana-reverse-gpt2-xsmall