seems to be using text-davinci-002-render-sha - not gpt-3.5-turbo - when you select ChatGPT 3.5 as the model?

I have some code that makes API calls to ChatGPT using gpt-3.5-turbo as the model, however, sometimes, for convenience purposes, I’ll test prompts on chatgpt[dot]com vs through my API. Since I’m using gpt-3.5-turbo in the API calls it stands to reason that the model I should be using on chatgpt[dot]com is ChatGPT 3.5, however, when I select that, the GET parameter that’s added is set to text-davinci-002-render-sha. According to ­platform[dot]openid[dot]com/docs/models/gpt-base “most customers should use GPT-3.5 or GPT-4” and yet chatgpt[dot]com doesn’t seem to be doing so when ChatGPT 3.5 is selected from the drop down menu at the top left? Why would GPT-3.5 be using a model that says shouldn’t be used? Seems like a bug?