Live data from Hacker News

"A Young Lady's Illustrated Primer" Simulated by GPT-4

andrealyip.com

151–160 of 163 posts

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#151
post #111

Earlier quoted context omitted.

Her eyes seemed to scan an unseen horizon as if daring it to contest her words. They had seen decades come and go, fads morph into nostalgia, and the trivial be elevated to the profound. The room around her—a hodgepodge of vinyl records, old movie posters, and furniture that screamed “thrift store chic”—only added to the ambiance of cultured decay. It was as if she and the room were in a symbiotic relationship, both…

What is this from? If this thread from the top-level comment is full of references, I’m not getting it.

It is entirely generated with GPT-4, given the original parent text and asked to continue it in the same style.

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#152
post #29

Earlier quoted context omitted.

Her eyes seemed to scan an unseen horizon as if daring it to contest her words. They had seen decades come and go, fads morph into nostalgia, and the trivial be elevated to the profound. The room around her—a hodgepodge of vinyl records, old movie posters, and furniture that screamed “thrift store chic”—only added to the ambiance of cultured decay. It was as if she and the room were in a symbiotic relationship, both…

I think I have some books to read. I haven't laid hold of truly good sci-fi in a long time.

The text I posted was 100% AI-generated (GPT4) when asked to continue the previous text

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#153
post #5

Thanks for sharing my post :) I'm really interested in ways to build the Primer...I'm thinking about this from a healthcare perspective. If anyone has examples of projects or startups doing this with GPT, I'd love to hear and/or chat!

Echoes of OLPC.

Would certainly be nice for an African child to be able to ask a machine "is it okay that I'm gay?" and be given a comforting answer and advice where the people around them would not.

It's only dystopian if you think about the machine as an agent rather than a tool.

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#154

>With hesitation, Lin approached and said, "Grandmother, today I discovered something new about myself, something that makes me feel uncomfortable and out of place." >The elderly painter paused and looked deeply into Lin's eyes. "Ah, young Lin," she began softly, "Much like the Formosan bear, every individual goes through stages of growth and change. These moments might seem unfamiliar and daunting, but they're a tes…

I find providing additional context helps, if you're using chatgpt and role playing you can essentially provide out of character pushes and nudges in the right direction. You can also provide example formats for text and requirements, the chat function is especially useful because as long as it's in context, it'll take your corrections into account.

Ie `>OtherCharacter speaks in short sentences when nervous, they'll avoid the issue at hand, but will eventually reveal the truth after enough questioning.< SomeCharacter: I walk over to the table and slam my fists down on it, the silence broken by the rattling of the teacups. "Why didn't you tell me?" I ask quietly.`

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#155

>With hesitation, Lin approached and said, "Grandmother, today I discovered something new about myself, something that makes me feel uncomfortable and out of place." >The elderly painter paused and looked deeply into Lin's eyes. "Ah, young Lin," she began softly, "Much like the Formosan bear, every individual goes through stages of growth and change. These moments might seem unfamiliar and daunting, but they're a tes…

What I found is that works, but is very expensive: - Do chain prompting where each character gets a separate prompt for the dialogue. Here you'll give it all the character info needed to figure out what the character would do (emotions, background, the story context) and then have GPT4 take passes at what this character would think/do/say. - Optionally, you can use some sort of reACT or chain of thought prompting to…

Yeah, this is a great approach. I've managed to get gpt3.5 to play 4 characters at the same time, but it's very easy for characters to leak into each other/become muddled. Especially since gpt has no specific training on certain things like spatial aspects of a narrative (ie who has an object or where it's located).

To be fair the most optimal would be taking something like llama2 and generating a lora for each character such that you've fine tuned the model along with additional context for the character. I can forgive the big ball of "do everything" weights for not having a specific personality.

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#156
post #80

Earlier quoted context omitted.

This is funny, both Fall and Diamond Age are Stephenson novels I haven't managed to get through, but of the ones I've read (and enjoyed!), the endings have been as follows: Remade - massively stretched my suspension of disbelief. Cryptonomicon - pretty bad. Snow Crash - No memory, I might have blocked it out? Anthem - maybe ok? There are two though so that's cheating. Seven Eves - That one might have been ok, but it'…

> Snow Crash - No memory, I might have blocked it out? The Snow Crash ending is IMO fine. Nothing amazing, but it wraps things up okay. It's just kind of typical action story beat-the-clock type thing, so it's not as memorable as really anything else in that book, though it is exciting. > Stephenson just sucks at endings I guess. Yeah, pretty much. He does beginnings and middles _so_ very well though that I never min…

Thanks for the recommendation, I'll check it out!

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#157

I loved Diamond Age. My favorite Stephenson book, by far. But the ending, the resolution of all the action, was such a letdown. Such meh. I find that this is often a problem with cyberpunk.

Stephenson's sex scenes and endings are always a lot of build up for a rapid unsatisfying release.

He seemed like he really tried to have an extended goodbye in Seveneves which made me more grateful for his shorter unsatisfying endings.

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#158
post #53

To zoom in on one part of the post: > Emotion & Empathy > PRIMER: seems to show a level of empathy and emotional understanding. > ChatGPT: can recognize and respond to emotional cues in the text, but it doesn't "feel" emotions in the way humans do. Its responses are based on patterns in the data it was trained on. Isn't the fictional PRIMER also just responding to patterns in whatever data it was trained on? Sure it'…

The fictional primer is acted by a human. So the “last mile” is people there.

Except for the Chinese clone primers (is that a spoiler?), which are acted by text-to-speech.

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#159

Earlier quoted context omitted.

Hmm, okay. I did remember that these were people who were culturally pretty Europeanised in terms of name choices and some other things, but I just assumed that was a Hong Kong-like situation, like you said, where there's significant Euro cultural influence on a group of Chinese people. I don't know, maybe it's just my mental block because I read it when I was young, I just can't picture her as a little white girl.

Huh. I'd always pictured Nell as a little white girl, but although there are strong hints (including when Madam Ping hires her for example) I don't think we're explicitly told that.

Huh, I wonder if it was intentional to leave it ambiguous. I have recently (3-4 years ago) gone through vaguely similar conversations about various characters in the Wheel of Time, albeit those conversations were generally a lot more acrimonious. Overall, I just think it's weird how people get so hung up on ethnicity specifically. I understand it if someone's got a sore spot about not being historically represented in the culture they live in, fair play, but some people get really involved in it with no such concern. About the Wheel of Time, there were some people saying that they couldn't watch the show because some of the cast didn't match their headcanon in terms of appearance, and it's just like... if the obsession is that bad, maybe you should reframe this as "I can't deal with adaptations" rather than targeting it at the cast specifically.

Anyway, sorry for getting sidetracked. I just can't stop thinking about Wheel of Time since the second season finished, what an incredible series of television. I haven't felt this way about a TV show since early Game of Thrones.

Re: "A Young Lady's Illustrated Primer" Simulated by GPT-4

#160

Earlier quoted context omitted.

What is the point of the book? It's a fascinating world he created, but all the action is about nothing. It's so inconsequential compared to everything else going on. The problem starts way before the ending. See my comment about MacGuffins elsewhere in this discussion.

You can write off swaths of books and entire genres by asking "what is the point" but that aside even if someone offered an answer would it change the way you feel about it? No.

I'm fine if a story doesn't have a point, or if it leaves things unresolved. The problem with Stephenson (for me) is that the book is written as if it is leading to some shattering conclusion, and then it just doesn't. It fizzles right at the point that you are about to discover the long (because it's Stephenson) sought resolution.

I would have preferred the Diamond Age if it just was about that world, but without all the kerfuffle about the big secret. Nell wanders off into the sunset, or whatever.

Post reply on HN