Great article Charles! There's some great nuggets in here, but calling out the limited size of training data you can currently use for you own GPT is very surprising. I was wondering if you've done any work creating GPTs with numerical datasets say a CSV of historical weather data for a city. I'd be interested to see if ChatGPT works better or worse with that format when creating your own GPT. Thanks for the article!
The GPTs are weird, in that you can build a lot of complexity into them, but they run on ChatGPT's platform (the one you pay $20/month to access). The limits seem to be from their using an older version of GPT-4. But there's scant documentation, especially about limits, so knowledge comes mostly from trial and error. I would definitely encourage you to play around with your ideas!
Great article Charles! There's some great nuggets in here, but calling out the limited size of training data you can currently use for you own GPT is very surprising. I was wondering if you've done any work creating GPTs with numerical datasets say a CSV of historical weather data for a city. I'd be interested to see if ChatGPT works better or worse with that format when creating your own GPT. Thanks for the article!
The GPTs are weird, in that you can build a lot of complexity into them, but they run on ChatGPT's platform (the one you pay $20/month to access). The limits seem to be from their using an older version of GPT-4. But there's scant documentation, especially about limits, so knowledge comes mostly from trial and error. I would definitely encourage you to play around with your ideas!