You called my post saying "no logic is happening" an oversimplification so, you pretty much did my friend. If you don't understand how the 'LMM's work and don't understand how a word prediction engine can at times appear to follow logic I don't know what to tell you
The chat only proves definetly that AI has a bias and that it lacks the ability to discuss abstract subjects that involve logic as it was trained using data from users that are unable to reach such levels of cognition.
Why would I assume my data was used to train the AI? I've had very little online footprint outside of here for decades.
So no I didn't see that as calling me stupid, I saw that conclusion as you exposing your ignorance about the inner workings of LLMs which you continue to do. If it massages your ego to insinuate your mental superiority while discussing topics you have limited understanding of by all means continue
Regarding my other point, you continue to prove it. You literally don't understand the architecture if you think it's the quality of training data that has anything to do with the system's inability to engage in logic. But you'll keep overrating yourself (like somehow thinking it should be a big deal that you made a poor attempt at a veiled insult), narcissism is big in conspiracy circles these days
You called my post saying "no logic is happening" an oversimplification so, you pretty much did my friend. If you don't understand how the 'LMM's work and don't understand how a word prediction engine can at times appear to follow logic I don't know what to tell you
I called you stupid and you've never noticed kek
Why would I assume my data was used to train the AI? I've had very little online footprint outside of here for decades.
So no I didn't see that as calling me stupid, I saw that conclusion as you exposing your ignorance about the inner workings of LLMs which you continue to do. If it massages your ego to insinuate your mental superiority while discussing topics you have limited understanding of by all means continue
It doesn't matter what you did online. The people that know you gave away your data. Parallel profiling is a thing.
Regarding your other point, LOL.
Parallel profiling was not used to train "LMM"s.
Regarding my other point, you continue to prove it. You literally don't understand the architecture if you think it's the quality of training data that has anything to do with the system's inability to engage in logic. But you'll keep overrating yourself (like somehow thinking it should be a big deal that you made a poor attempt at a veiled insult), narcissism is big in conspiracy circles these days