That’s an informative result. Your completions of 天 and ヤ are nothing like mine. My experiments never produced pinyin or any other phonetic transcriptions like Tiān or yā.
By the way, these experiments used text-davinci-003 via OpenAI’s playground. I don’t know how to access davinci-instruct-beta.
In the dropdown in the playground, you won’t see “davinci-instruct-beta” listed. You have to click on the “Show more models” link, then it appears. It’s by far the most interesting model to explore when it comes to these “unspeakable (sic) tokens”.
That’s an informative result. Your completions of 天 and ヤ are nothing like mine. My experiments never produced pinyin or any other phonetic transcriptions like Tiān or yā.
By the way, these experiments used
text-davinci-003via OpenAI’s playground. I don’t know how to accessdavinci-instruct-beta.In the dropdown in the playground, you won’t see “davinci-instruct-beta” listed. You have to click on the “Show more models” link, then it appears. It’s by far the most interesting model to explore when it comes to these “unspeakable (sic) tokens”.