GPT-4 Turbo is refusing to follow code refactoring instructions, and claims that it is due to a “platform limit”, which I assume is the completion token limit of 4096. Giving it a code file that clocks in at a whopping 1,920 tokens, it refuses to perform a refactoring operation as instructed, and i…

GPT-4 Turbo refusing to follow instructions

aronweiler December 7, 2023, 9:21pm 2

Additionally- this functionality worked in GPT-3.5 Turbo, and still works now.

JSON Mode with GPT-4 turbo stops after 1050 token

Topic		Replies	Views
Please stop anthropomorphizing GPT! Community gpt-4 , chatgpt	23	3030	December 17, 2023
Custom chatbot says that it's developed by OpenAI API gpt-4	33	1963	April 2, 2024
Ethics: Remove default fake emotions from ChatGPT API	29	5974	December 17, 2023
Why I Think GPT Is Now Lazy Community gpt-4 , chatgpt	30	18381	February 6, 2024
ChatGPT's Role Reversal? - When AI Asks Users to Take the Wheel Prompting gpt-4 , chatgpt-plugin	2	1383	January 6, 2024