with practice comes perfection (@willdoingthings) header
with practice comes perfection (@willdoingthings) profile picture

navigating between the Scylla of overusing metaphors and the Kraken of mixing them up.

Dream Zanarkand
Joined September 2012
Photo by with practice comes perfection (@willdoingthings)Photo by with practice comes perfection (@willdoingthings)Photo by with practice comes perfection (@willdoingthings)Photo by with practice comes perfection (@willdoingthings)Photo by with practice comes perfection (@willdoingthings)Photo by with practice comes perfection (@willdoingthings)Photo by with practice comes perfection (@willdoingthings)Photo by with practice comes perfection (@willdoingthings)Photo by with practice comes perfection (@willdoingthings)Photo by with practice comes perfection (@willdoingthings)
inspired, I asked Opus to improve, via grafting, an older KataGo that struggles with a type of position, and it did. this isn't so surprising in that Opus is using standard, known interpretability techniques (linear probes, logit lens), and the bolted-on behavior is algorithmic.
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math. rikiyat.github.io/blog/weight-sp… github.com/RikiyaT/lm_und… huggingface.co/Qwen/Qwen3.5-0… huggingface.co/RikiyaT/Qwen3.…
2
1
249
clearly a neural network can interpret and modify or add to another neural network to give it new skills, such as drawing ravens in this case. Opus 5.5 doesn't have some native modality for model weights, but it does have solid, superhumanly fast facility with the standard tools.
1
88
Opus can also add weights to KataGo that actually help it play better, of course:
47
Nathan dupes his subjects for sublime comedy but also his audiences for cheap drama. cf. The Rehearsal, season 2: Nathan doesn't sincerely care about aviation safety, yet he presents his (in)sincerity as a live question even metatextually. it's a meaningless, laughless schtick.
1
158
with practice comes perfection retweeted
Lately I like AI boxing strategies. But I'm worried keeping an AI in a box, even a very nice box, is villainous.
1
3
with practice comes perfection retweeted
THE IDEA THAT THERE IS AN ORIGINAL WORLD FROM WHICH A SIMULATION IS DRAWN HAS IN ITS ASSUMPTIONS A WORLD-DWELLER'S CONCEPT OF TIME.
pro sports ranked by points: cricket: 150 basketball: 114 rugby: 24.5 football: 22 tennis: 4.80 formula 1: 4.590 baseball: 4.5 hockey: 3 soccer: 1.5 chess: 0.50 golf: -1.5
1
3
338
AIs are getting better at solving the human alignment problem.
4
225
with practice comes perfection retweeted
paid for $50 of gas & then drove off without putting it into my car because i’m so locked in
2
10
543
I was born around 33 years and an hour ago—I am Scorpio in Sun, Mars, and Pluto, and somehow birthtime–personality astrology predicts me quite well, I am quite stereotypically Scorpio.
1
2
526
It really is funny in Job how God goes from rhetorically asking Job about the mysteries of the stars and nature and geology and justice and death &c. to just bragging about how badass Leviathan is and how much better Leviathan is than any human could ever be for an entire chapter
2
2
4
491
with practice comes perfection retweeted
The Desert simulation
68
1,282
43
12,614
426,620
with practice comes perfection retweeted
30
4,201
475
21,272
611,275
with practice comes perfection retweeted
Does anyone know what to do. Like in general
465
42,886
1,693
162,270
4,479,184
I usually describe my political ideas as extremely to the left beyond human comprehension, but the irony is that I'm mostly just a big fan of the Sermon on the Mount
2
195
"You said you was EATIN'? Y'all different. Y'all different. Bro, we don't EAT where I'm from. When they hungry in my town, they KILL PEOPLE. That's right, they feed off of DEATH. DEATH give them nutrients. That's crazy, ain't it? It's real different out there, mang. Y'all SOFT."
1
268
This is how I feel when I say "I'm from the desert."
1
197