Greg Clarke

Mastodon: @greg@clar.ke

  • 52 Posts
  • 224 Comments
Joined 2 years ago
cake
Cake day: November 9th, 2022

help-circle
















  • That’s my point, if the model returns a hallucinated source you can probably disregard it’s output. But if the model provides an accurate source you can verify it’s output. Depending on the information you’re researching, this approach can be much quicker than using Google. Out of interest, have you experienced source hallucinations on ChatGPT recently (last few weeks)? I have not experienced source hallucinations in a long time.











  • Yes of course I’m asserting that. While the performance of LLMs may be plateauing, the cost, context window, and efficiency is still getting much better. When you chat with a modern chat bot it’s not just sending your input to an LLM like the first public version of ChatGPT. Nowadays a single chat bot response may require many LLM requests along with other techniques to mitigate the deficiencies of LLMs. Just ask the free version of ChatGPT a question that requires some calculation and you’ll have a better understanding of what’s going on and the direction of the industry.