☡ pfp

☡

@stultulo

23 Following
72 Followers
☡ pfp
☡
@stultulo
khdjysbfshfsjtstjsjts
1 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
Maybe we just need better satirists, because whatever your satire’s trying to say, the audience only ever hears, “Don’t think about elephants.”
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
It’s fucked up that successful satire is always successful once it’s taken at face value, so you need to satirize the satire.
0 reply
0 recast
3 reactions
☡ pfp
☡
@stultulo
I was either thinking about the “Bad Blood” section of Rimbaud’s A Season in Hell at the time, or about writing a parody of Fight Club.
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
laughing remembering the first line to a (satirical) novel i once envisioned: “My father taught me to be weak, and my mother taught me to be cruel.”
0 reply
0 recast
4 reactions
☡ pfp
☡
@stultulo
“All poets, even the most avant-garde, need a father. But these poets were meant to be orphans. He never came back.” Excerpt From The Savage Detectives Roberto Bolaño This material may be protected by copyright.
0 reply
0 recast
2 reactions
☡ pfp
☡
@stultulo
awoken after strange dreams
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
i will ENDURE, i’ll find THE solution to whatever problem i SEEM to have determined MIGHT exist
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
The plight of the stultuloj and their stupid fucking word games played against moody LLM sampling parameters. All for for no tangible reason. Out of boredom. Just because.
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
The funny thing is that I think this is directionally correct, you’d want a newer model to sound more like GPT-4 at a higher temperature. But in the olden days, imo, setting GPT-4 to temperature: 1.2 was the sweet spot. Which is still like setting GPT-4.1 somewhere low, between 0.2–0.25, in my view. In other words, it’s cooked. GPT-4.1 is. And it’s hard to say whether you’re better off lowering the temperature (in the API) or doing a system message (more in like with ChatGPT). There’s a weird alchemy to it. I’m tired of this game.
1 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
incredibly: to sound like gpt-4-1106-preview @ temperature: 1.0, I have to set gpt-4.1 wayyy down to 0.2 but 4 doesn’t even begin to sound like 4.1 @ temperature: 0.7(!) until you set it way up to 1.5(!) ….which is also where GPT-4 tends to do the honorable thing & start speaking gibberish instead.
1 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
maybe all i need to do…I read somewhere a lot time ago that ChatGPT was set to 0.7 temperature. Based on a few repeated tests in the API, 0.7 still seems about right. The temperature is definitely less than 1, anyway. This doesn’t fix the problem tho, it reduces it by 30%.
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
https://warpcast.com/stultulo/0x2d1b4215
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
https://warpcast.com/stultulo/0xe226951b
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
Insane hell world. I'll just do a DPO fine-tune that literally uses GPT-4 Turbo responses as the "correct" examples and GPT-4.1 responses as the "incorrect" examples. I have enough saved chats that I could throw one together real quick. Why not. It doesn't have to be great, it just has to show improvement.
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
cause of death: RLHF
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
Modern LLMs also need an intro and four section headers to express the equivalent of GPT-4 Turbo's first two paragraphs. All of these use the same chat history and no system message. This is Not Good. Anything that was once remotely insightful, even just from the perspective of an "augmented journal" has been laid to waste. But on the other hand, they've succeeded at making sure that LLMs can attempt to debug your Github repo and only fail half the time. It is what it is.
0 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
This screenshot is how LLMs used to talk by default, from a GPT-4 Turbo convo from 2023. It’s talking about the antepenultimate chapter in the final book (it was a trilogy back then). In other words—the very ending, where the tension is highest, bc despite being a hybrid genre, the series is a spy thriller at heart. Modern LLMs, including Claude 4, will begin their response by saying of this chapter, “This is…” ❌ —“…an outstanding, high-stakes climax, loaded with personal and philosophical fallout…” —“...a powerful, loaded scenario—one that delivers intense catharsis, profound discomfort, and, crucially, major moral ambiguity. It aligns beautifully with the Nietzschean necessity/beauty dilemma…” —“...a stunningly tense, subversive, and emotionally loaded conclusion for your narrative—and it delivers magnificently on…” —“…absolutely devastating and brilliant…” vs. ✅ —“The denouement you describe here hinges on Gemini's moral calculus…”
1 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
I tried both on the console (with extended reasoning, other attempts were without) and they're more sane now. Both are less likely to say that every other idea is "absolutely stunning" or "powerfully brilliant" or "richly layered and morally complex." In fact, the phrasing and insight is almost on par with GPT-4 Turbo. As in the November 2023 model 🫤 I really wish one of these big AI companies would give a shit about improving the models at things other than coding and telling the user that they're a fucking genius for adding a miniscule wrinkle in a character's personality. Words like "genius" and "brilliant" used to mean something.
1 reply
0 recast
1 reaction
☡ pfp
☡
@stultulo
on the slackline of derking my shmerk rn with these AIs
1 reply
0 recast
1 reaction