viktor-ferenczi

4 ms·

1. 1.▲
   by viktor-ferenczi 3y ago

   After reading all these, I've landed at the following conclusion: Regardless of its architecture there is only a finite amount of information the language model can work with at any given time. It depends on the task at hand which way
2. 2.▲
   by viktor-ferenczi 3y ago

   Exactly
3. 3.▲
   by viktor-ferenczi 3y ago

   That's part of the reason why we need LLMs to run locally (on our own or rented infrastructure). Another reason is protecting the company IP. None of the medium/large corporations want their IP to be leaked to AI providers.