Upgrade your account to a Community Supporter account and remove most of the site ads.

Reply to thread

Message: <blockquote data-quote="RareBreed" data-source="post: 9089452" data-attributes="member: 6945590">No, this is not just a lack of documentation.  The model layers have become so deep, that the data scientists developing them can no longer explain how the program is able to solve the problem in the way that it does.  Hence, <a href="https://www.scientificamerican.com/podcast/episode/what-you-need-to-know-about-gpt-4/" target="_blank">emergent behavior</a> in these LLMs.  An example of this is that GPT-4 was not trained to learn how to perform arithmetic, but it can do it anyway.That is what makes these LLM's scary.  They have become so complex with so many "moving parts" that while we have a general understanding, we can't explain them.  It's not much different than how we know how the neurons in our brain fire and how they have limiters that can squelch the firing (in machine learning, the analog to this are called Activation Functions, that convert the linear dot product calculated slope with a weighted Y-intercept bias into non-linear functions....these are usually implemented as ReLU or some kind of sigmoid/softmax function).  But even though we have a good idea how individual neurons work, we have zero clue how the mass interaction of them creates thoughts, emotions, memory or our subjective sense of a "mind".While it is true that the model architecture and the data set trained on are proprietary, that's not so people can figure out how they work.  The scientists who created some of the new generation Transformers don't even know how they are doing what they are doing.  Some customers are demanding "explainability" and honestly, they dont know how it actually works, and they are literally called black boxes.</blockquote>

Verification