Curated Resource ( ? )

Running Python micro-benchmarks using the ChatGPT Code Interpreter alpha

my notes ( ? )

"Code Interpreter mode ... [allows] ChatGPT ... generate Python code and execute it in a restricted sandbox" - no network access, can access pre-approved libraries.

Author provides an example where ChatGPT's code didn't initially work, but without prompting it diagnosed the problem, fixed it and ran it again, leading him to offer "another analogy for large language models ... like having a coding intern...:

  • memorized all of the world’s public coding documentation, though they stopped reading in September 2021.
  • unbelievably fast
  • If their code produces an error they can read the error message and attempt to update their code to fix it.
  • won’t argue with you... not necessarily a good thing...
  • respond incredibly well to feedback
  • You have to know how to prompt them
  • You have to closely review EVERYTHING they do.
  • They work for free."

Read the Full Post

The above notes were curated from the full post simonwillison.net/2023/Apr/12/code-interpreter/.

Related reading

More Stuff I Like

More Stuff tagged python , chatgpt , llm , plugin , code interpreter

Cookies disclaimer

MyHub.ai saves very few cookies onto your device: we need some to monitor site traffic using Google Analytics, while another protects you from a cross-site request forgeries. Nevertheless, you can disable the usage of cookies by changing the settings of your browser. By browsing our website without changing the browser settings, you grant us permission to store that information on your device. More details in our Privacy Policy.