Comment

Comment on How Googlers cracked an SF rival's tech model with a single word | A research team from the tech giant got ChatGPT to spit out its private training data

<- View Parent

Sanctus@lemmy.world ⁨1⁩ ⁨year⁩ ago

A breakdown in the weighting system is the most probable. Don’t get me wrong I am not an AI engineer or scientists, just a regular cs bachelor. So my reply probably won’t be as detailed or low level as your’s. But I would look at what is going on with whatever algorithm determines the weighting. I don’t know if LLMs restructure the weighting for the next most probably word, or if its like a weighted drop table.

source

Sort:hotnew top

luthis@lemmy.nz ⁨1⁩ ⁨year⁩ ago
Hate to break it to you, but you’re more qualified than me!

I only did a Coursera cert in machine learning.

source
- Modva@lemmy.world ⁨1⁩ ⁨year⁩ ago
  My fun guesswork here is that I don’t think the neural net weights change during querying, only during training. Otherwise the models could be permanently damaged by users.
  
  source
  - luthis@lemmy.nz ⁨1⁩ ⁨year⁩ ago
    The neural net doesn’t change of course, but previous text is used as context for the next generation.
    
    source