• 1 Post
  • 128 Comments
Joined 1 year ago
cake
Cake day: June 4th, 2023

help-circle













  • LLMs cannot:

    • Tell fact from fiction
    • Accurately recall data from its training set
    • Count

    LLMs can

    • Translate
    • Get the general vibe of a text (sentiment analysis)
    • Generate plausible text

    Semantics aside, they’re very different skills that require different setups to accomplish. Just because counting is an easier task than analysing text for humans, doesn’t mean it’s the same it’s the same for a LLM. You can’t use that as evidence for its inability to do the “harder” tasks.






  • I would caution against using these numbers for any Calorie-in/out calculations. Even if they were 100% accurate, it still doesn’t take into account anything that happens outside of that machine.

    Example: There’s something we call “non-exercise activity thermogenesis” (NEAT). This includes lots of things you do without thinking about it, like fidgeting, tapping your feet while sitting, or pacing around the room. In some people, NEAT can decrease significantly after exercise, which then negates much of the Calories burned.