brubsby (@brubsby) header
brubsby (@brubsby) profile picture

computer user /ˈbrʌbzˌbi/ https://xitter.cf/t.co/mQtYsGU5h5

Joined March 2014
Photo by brubsby (@brubsby)Photo by brubsby (@brubsby)Photo by brubsby (@brubsby)Photo by brubsby (@brubsby)Photo by brubsby (@brubsby)Photo by brubsby (@brubsby)Photo by brubsby (@brubsby)Photo by brubsby (@brubsby)Photo by brubsby (@brubsby)Photo by brubsby (@brubsby)
a noodle so nice, they dan'd it twice
1
175
brubsby retweeted
Imagine if Erdos were still alive with a GPT-5.5 Pro subscription and amped up on amphetamines.
38
74
22
1,784
81,268
Ok, so I learned that these are called phonesthemes :)
Today I learned that in English: words starting with "gl-" often relate to light: gleam, glint, glance, glare, glaze, glimpse, glitter, gloss, glower, glisten, glaze, glitz, glazed. words starting with "sn-" often relate to nose: sniff, sneeze, snuff, snore, snort, snot.
100
7,686
265
57,945
1,581,475
explaining Maxwell’s demon to a border collie
This is oddly satisfying 🥹
12
101
1,594
169,368
brubsby retweeted
this is what it looks like to traverse all the zip codes sequentially:
36
176
24
4,859
430,432
グラフ上で点をグリグリ動かして母音を発音させるやつをp5.jsで作ったことがあります
36
1,122
107
5,897
316,593
One common issue with personalization in all LLMs is how distracting memory seems to be for the models. A single question from 2 months ago about some topic can keep coming up as some kind of a deep interest of mine with undue mentions in perpetuity. Some kind of trying too hard.
1,716
1,043
466
20,839
2,796,135
Replying to @zanehkoch
When I was on the ISS for my nearly year long mission, there was a telomere experiment comparing my telomeres to my earth baseline and my twin brother as a control. Hypothesis was they would get damaged and worse due to the environment. Turns out they got better. Initially NASA thought maybe it was due to exercise and diet. After I returned we learned JAXA had a telomere experiment on some small worms the same time I was there. Their telomeres got better too. Never saw the worms doing any exercise. After further study determined it was the radiation.
142
405
84
8,413
1,284,032
if your company uses workday for its ATS i am simply Not Applying
157
my kingdom for an llm that says "o ya oops" instead of "You are absolutely correct!"
69
trying to heal the political divide: dang I could really go for some stone soup right now
78
forbidden post:
1
75
uh oh chat
We estimate that, on our tasks, Claude Opus 4.5 has a 50%-time horizon of around 4 hrs 49 mins (95% confidence interval of 1 hr 49 mins to 20 hrs 25 mins). While we're still working through evaluations for other recent models, this is our highest published time horizon to date.
1
1
99
I thought the hackernews types foaming at the mouth against github's new ui were overreacting but it's so slow for some of my PRs that chrome actually just tries to quit and kill the page
2
87
my beautiful wife gave birth to a happy and healthy baby boy, and i get to spend every day of the rest of my life loving them. cold shredded cheese, bacon, and mayo sandwich.
8
239
girl i'm bored let's burp you halfway through the bottle so you don't spit up
1
580
pressing buttons on the extreme corners of my phone with my nose >>>>>>>
1
70
every year or so since chatgpt i check out how good llm codegen is, this year i tried getting the frontier models to do some boilerplate for a CUDA program and they all get stuck in an infinite loop trying to fix all the c++ dependencies lmao
1
69
llm cli environments are fun though cause you get to unleash a little computer homunculus on your problem and shut the door, then you come back later and they've completely destroyed your wizard tower
56
having thinkpad withdrawals after giving back my work thinkpad. what are we thinking chat, x1 carbon gen 8 with arch?
1
2
148
found an IT guy on marketplace that has stacks of thinkpads from businesses retiring them and now we're friends and I got a T590 for $100
1
62