Wikidepia/IndoGPT2-small

Thanks to Tensorflow Research Cloud for providing free V3-8 TPU . GPT2-small is trained on:OSCAR Corpus, Wikipedia, Wikipedia and OpenSubtitles . Using GPT-Neo from EleutherAI. The model is based on the GPT3-like model used in Indonesia . The model was trained on OSCAR Corpus Corpus and Wikipedia's Open Subtitles. It is now being used to train the model on GPT 2-small GPT 3-like GPT Neo-GPT2 . The data was used to build a model of a model that was trained using GPT Neoscript. The code is now available to download,