Outdoorsman, musician, philosopher, ✝️. Western explorer in the eastern concrete jungle. Daring all that becomes a man.
Alas! New York
Joined July 2023
- Tweets5.8K
- Following854
- Followers313
- Likes35.7K
Pining for sagebrush and shortgrass and yellow aspen and cottonwoods and small dry windswept snowdrifts
How to find smart borderers
Smart borderers where
Borderer women high income
Borderers high openness
Borderer-german intermarriage Indiana
Folksy homey ethnicities
Homemakers who like hiking
America musical genetics map
Feminine women who like mountains and singing
SomeBrashAtom retweeted
4) While this is going on, we probably get the bioweapon version of Black October. A new dangerous pathogen pops, and then possibly another a few weeks later, and so on.
At first, it's not necessarily clear they're distinct pathogens, or AI-designed.
Overproduction of people commanding elite resources, or aspiring to do so
Underproduction of people who deserve to
Modern Mathematics has been this way! Most numbers are uncomputable and most functions are nowhere continuous. The more interesting question is, what is the local curvature of math which leads to useful insights about our world? I expect it to is positive but quite small
increasingly concerned that ai is revealing the mathematical world as negatively curved. instead of everyone wandering about in idea-space and clustering around the same famous conjectures (local positive curvature), ai is pushing out the boundaries rapidly and revealing that the space of mathematical ideas is ~hyperbolic (negative curvature). though this means there's lots left to discover, it also means that small separations between people/agents will lead to wandering alone.
we may be emerging into an idea-space where there's so much to see and do but everyone is alone in a private heaven, unable to share their subtle joys with others...
I love seasonal decorations. Living in a house decorated for Halloween, or for any season, anchors me. It makes me feel like I'm living in a real place instead of a fake temporary holding zone for people who haven't yet managed to get married and form a household
SomeBrashAtom retweeted
If you can't attend the Avalon Exposition but want to come meet speakers and friends coming in from out of town, come join us for our Halloween party! There will be duels, drinks, dark stories, and a costume contest with a 1k cash prize!
Link in the comments
SomeBrashAtom retweeted
We shouldn't work on alignment or interpretability. The "good outcome" here is we build a benevolent God to trap us in utilitarian mouse utopia forever. We simply should not advance this technology at all.
interpretability might be the most important scientific problem in the world right now, as I don't think we can solve alignment if we don't know what we're building. I think I have good news about it. I co-authored early work in mechanistic interpretability at OpenAI and came away pessimistic, because neural networks appeared to fold into arbitrary tangled knots. years later, I started openai's first large-scale automated interpretability project, finding about 2% of neurons legible with the primitive tools we had
with modern agentic tools and a much closer look, the picture looks totally different. the vast majority in models up to trillions of parameters are legible, and their circuits beautiful. many famous examples of tangled neurons turned out to be misunderstood. The roadblocks I expected mostly aren't there. I now think ambitious interpretability will be solved by the time we need it most, and possibly quite soon, as automated scientists keep improving. it may be a bitter lesson for interpretability: the simple thing, at absurd scale, works
already-clean.goodfire.com/
SomeBrashAtom retweeted
This is good, this is a key part of how humans maintain morality and deontological reasoning is implicitly based on it, it is one of the basic cures to naive consequentialism. What is wrong with you?
