- API Key Registration: To use this chat program, you need an API key. You can register and manage your API keys at https://platform.openai.com/account/api-keys. ↗ to see which API’s you have access to you can check: https://platform.openai.com/account/rate-limits
- First Time Setup: First Time Setup: extract and run KevinGauthun_chat1example.exe. When you run the program for the first time, click ‘OK’ to create a new alias. Ignore the “Enter new alias” prompt due to a current bug. Enter your API key followed by a password of your choice. This password will be used to encrypt your API key using base64 encryption, a method that converts binary data into ASCII string format. You may be asked to re-enter your password. It will also create a api_keys.json file holding your alias and encrypted key wherever the exe is located.
- Subsequent Use: When you run the program again, select your previously defined alias. Enter the password you set up previously (even if the prompt says “Enter new password to encrypt API key”). This password is required to decrypt the base64 encoded API key. I plan to revise the prompts in a future update for clarity.
- Model Setup: By default its using the base gpt-3.5-turbo model but I included access to gpt-4 if you click on the model button in the top right. (as long as your api key/account has access to that model, it will switch over to it)
- Context Setup: Content/context represents the context of the chat. For example, you could set the context to “You are a junior program developer, programmed to be helpful with coding tasks and analysis.” Or “You are a executive chef at a 5 restaurant. Your goal is to provide unique dishes and recipes to be prepared”
- Temperature Adjustment: Temperature controls the precision of the responses. Lower values produce more focused and deterministic responses, while higher values yield more diverse and creative responses
- @ July 22, 2023 6:35 pm