swlabr
Only Bayes Can Judge Me
actually I found this ancient text online that helps explain everything:
Eliezer Monogatari, chapter 69:
On his mighty steed Oververbosity, Eliezer is fast approaching the cliff edge overlooking the dreaded Chasm of Unreadability. His desire to cram more words and IQfulness into his text has deafened him to the wails of the Ghost of Editors Past, begging him to at least paste into Google Docs and fix a few squigglies.
In a moment of rational brilliance, he casts Bayesian Reasoning and epiphanisationalises that he can Just Add Metainfo to every line, exponentially increasing his QLAWPW (quality of life adjusted words per word), making him The Greatest Author. His loyal mount leaps into the air over the abyss, buoyed by clouds of self-delusion.
Of course, he crashes and falls into the chasm. But what’s this? The miasma of unreadability… it’s more clouds of self-delusion generated by the cult of TREACLES! He rises, unscathed, head three sizes bigger than before.1
1: Scholars are divided on whether this is because the toxic miasma mutated his sense of self-worth or if he hit his head on the chasm floor and it swelled. We will never know.
Fake answer: at some point in rat history they thought that Armenian people were genetically superior and decided to take traditional Armenian surnames as their handles, but they both simultaneously and serendipitously bungled the spelling. The names should actually be “Yvian” (pronounced yiv-yahn) and “Iarwian” (pronounced ee-arh-wee-yahn)
Real answer: probably a tolkien elvish name idk
“Editing? Where we’re going, we don’t need editing!” - Absolutely not Eliezer. Too succinct.
In my pursuit of clarity and precision, the art of composition itself becomes a meticulous process of refining ideas. Each word is a deliberate stroke, contributing to the coherence of thought, rendering traditional editing superfluous. The absence of formal editing is a deliberate choice, reflecting the belief that the initial act of creation captures the most authentic expression of my intellectual journey.
Now that is what yud1 might say.
1: By Yud I mean a 99.9% accurate simulation of Yud using ChatGPT, which, according to rationalist doctrine, is the same thing.
More like… Project Awful
boom roasted
my reference point for this kind of extension is the one that changes “social justice” and “sjw” with “skeleton” and “skeleton warrior.” For example:
“sjws are taking over X” -> “skeleton warriors are taking over X”
Actually now that I’m typing this I hope there’s a good one for “woke”.
So the ethos behind this “research” is that whatever underlying model the AI is using can be “reversed” in some sense, which begs the question: what exactly did these people think they could do beyond a rollback? That they could beg the AI to stop being mean or something?
They were probably inspired by the blanka creation scene from the street fighter movie where they brainwash some guy by showing him video clips of bad stuff and then switch it to showing good stuff.