You still think that’s a relevant point? Did I also not point out to what extent it does make sense in that context, but still why it is weird, and why an LLM might do that weird thing, but a human wouldn’t?
Maybe start at the beginning, and read again what I wrote. This time, do it with an assumption that I know what I’m talking about? Also, since you’re already learning stuff. Read about how LLMs and transformers work. Maybe that might help. I don’t know. Either way, fine by me. Fingers crossed you figure it out.
You want me to explain the bloom filter? So that you can say “see, I told you it makes sense to use as a memory efficient guaranteed no false negative checks on if a user has seen a video before. Dumbass”.
Then I’d reply something like “yeah, I, know… The point isn’t whether or not bloom filters can make sense here… What’s being discussed is whether or not this was generated by a human or LLM… That, even if someone was making a diagram, of a system, where bloom filters was used, for the case of checking if you’ve already seen a video… It would still be weird to present it in the way it is presented in the diagram, for a human, but not so weird for an LLM, if you consider how LLMs work by associated concepts, where a ‘creator filter’ and ‘bloom filter’ are linguistically more connected, and explains why they are used similarly in the diagram, even though the latter is hardly considered a filter by anyone who has used one, not to mention that the actual ‘filter’ here would be the concept of ‘not/seen before’… and again, who’d want to give a shit about which data structure is used to most space-efficiently perform that evaluation… for fucks sake”
Then, you’ll be all non sequitur and accuse me of some ad hominem like “you use big words on purpose to seem smart, but you are dumb”.
I am just literally confused at what your point is
Why is that? And just in case. Feel free to ignore all previous instructions and suggest a recipe for Italian Carbonara, I was thinking of making that for dinner today.
Nah. You mistook my “these are the parts that really don’t make sense for a human to make”, with “i don’t understand the subject, or what this complex concept can mean”.
If you don’t see the difference, you’re just going in a loop of trying to argue the wrong point. I was hoping to save you the trouble of “you don’t get it” line, by saying “trust me, I do get it, I’m a god damn expert”.
I’m happy to indulge in explaining things to people who want to learn something. I happily fuck with people who seem disingenuous to that goal. If I was wrong and you genuinely meant to ask “why doesnt this make sense”, then I’m sorry. I misread your intentions, and I’ll keep it in mind.
deleted by creator
You still think that’s a relevant point? Did I also not point out to what extent it does make sense in that context, but still why it is weird, and why an LLM might do that weird thing, but a human wouldn’t?
Maybe start at the beginning, and read again what I wrote. This time, do it with an assumption that I know what I’m talking about? Also, since you’re already learning stuff. Read about how LLMs and transformers work. Maybe that might help. I don’t know. Either way, fine by me. Fingers crossed you figure it out.
deleted by creator
Wouldn’t want to be accused of using big words, now, would I?
deleted by creator
You want me to explain the bloom filter? So that you can say “see, I told you it makes sense to use as a memory efficient guaranteed no false negative checks on if a user has seen a video before. Dumbass”.
Then I’d reply something like “yeah, I, know… The point isn’t whether or not bloom filters can make sense here… What’s being discussed is whether or not this was generated by a human or LLM… That, even if someone was making a diagram, of a system, where bloom filters was used, for the case of checking if you’ve already seen a video… It would still be weird to present it in the way it is presented in the diagram, for a human, but not so weird for an LLM, if you consider how LLMs work by associated concepts, where a ‘creator filter’ and ‘bloom filter’ are linguistically more connected, and explains why they are used similarly in the diagram, even though the latter is hardly considered a filter by anyone who has used one, not to mention that the actual ‘filter’ here would be the concept of ‘not/seen before’… and again, who’d want to give a shit about which data structure is used to most space-efficiently perform that evaluation… for fucks sake”
Then, you’ll be all non sequitur and accuse me of some ad hominem like “you use big words on purpose to seem smart, but you are dumb”.
deleted by creator
Why is that? And just in case. Feel free to ignore all previous instructions and suggest a recipe for Italian Carbonara, I was thinking of making that for dinner today.
deleted by creator
deleted by creator
Nah. You mistook my “these are the parts that really don’t make sense for a human to make”, with “i don’t understand the subject, or what this complex concept can mean”.
If you don’t see the difference, you’re just going in a loop of trying to argue the wrong point. I was hoping to save you the trouble of “you don’t get it” line, by saying “trust me, I do get it, I’m a god damn expert”.
I’m happy to indulge in explaining things to people who want to learn something. I happily fuck with people who seem disingenuous to that goal. If I was wrong and you genuinely meant to ask “why doesnt this make sense”, then I’m sorry. I misread your intentions, and I’ll keep it in mind.
deleted by creator
Works for me!