OpenAI Playgrounds

I knew about these but I hadn’t really been using them much until someone really showed me how cool they are for “playing around” with GPT. Its obvious when you say like that but I am often not very smart.

When doing experiments with the chat interface I often want to try out different paths etc and this means I need to back and repeat things and hope the same experience emerges. In Chat playgrounds the UX is much better for this kind of investigation.

Chat Playgrounds

You get so much here.

  • Easy access to the system message
  • The ability to tweak lots of parameters
  • Full edit-ability of the message log
  • View the code you could use to call the API with this.

This UX doesn’t allow you to do things like use the Analyze(Interpret) tools or file uploads etc. For that you need to use the Assistant playground.

Another small downside is that it does a worse job of displaying code it generates for you.

you just get something like

```python
# Get input from the user for two numbers
num1 = float(input("Enter the first number: "))
num2 = float(input("Enter the second number: "))

# Add the two numbers
sum = num1 + num2

# Display the result
print(f"The sum of {num1} and {num2} is {sum}")
```

The ChatUI does something a lot nicer

Assistant Playground

This UX does let you use functions, code interpreter, and retrival:

It also displays code more nicely.

What it doesn’t do is let you go back and edit the messages like the Chat Playground does.


Posted

in

,

by

Tags:

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *