Currently a Platform Engineer Ex Datacenter Eng @ Google & Meta. Focus on Geopolitics, AI, Cyber Sec, Wardogs, OSS Posts and their opinions are my own.
Ankeny, IA
Joined November 2017
- Tweets9.1K
- Following888
- Followers308
- Likes40.2K
This is why @RobSandIA should prioritize clean drinking water for all Iowans because this guy has gone retarded.
Under Rob Sand,
Day 1) John Deere closes 2 plants
Day 2) Shooting at government building
Day 3) Shooting at McDonalds in Ankeny by illegal alien
Day 3) Democrat judge runs over someone, Rob Sand expected to pardon her
Day 4) 2 dead after multi county high speed chase
Day 4) Unemployment scam
But a temporary extra $1 for gas is the end of the world! Iowa libs are retarded for voting a transgender billionaire as our governor.
255... 256... 257? root. why?
I'm not even sounding the claxon for this it's just horrible.
Any text > 256chars is a valid root password, holy shit
sshhole.com/
3
108
A whole lotta people that don't live in my state dooming because we voted for Rob. We've been a swing state for a long time.
Just because your candidate lost doesn't mean that we still can't work together to achieve the things that both parties want, accountability, affordability, and right to choose and bear arms.
3
55
#factorio deathworld from lastnight (i never play with biters) so I teamed up with 8-12 other brave souls and after 16 hours we we're consumed by Shai-Hulud.
1
2
127
When @POTUS hires Nikki as the press secretary next week on the eve of invasion for Iran.
62
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.…
1
91
Today for no reason at all:
Is your fridge running? YOU BETTER CATCH IT (and patch it as soon as humanly possible)
I have to be like the top 1% of listeners of @Sleep_Token at this point. Can't wait for a tour in the midwest so I can see them live.
77