Kip Kaehler
140 posts
Kip Kaehler
@KipKaehler
Habitual line stepper.
San Francisco Katılım Aralık 2011
469 Takip Edilen93 Takipçiler
Gavin Newsom please spend $100 million to send cougars to my neighborhood
Christopher F. Rufo ⚔️@christopherrufo
The craziest thing about Newsom's $100 million wildlife bridge is that it will allow cougars, an apex predator, direct access into a suburban neighborhood filled with pets, children, and the elderly. It's like the radical environmentalist version of The Purge.
English
@emollick Isn’t this debate irrelevant? They’ve proven that MLA+MOE+RL w/o proprietary training sets outperform at trivial cost. Those ideas are everyone’s to take now, nobody needs to use the actual weights when training is this cheap. If not deepseek, everyone will use some derivative.
English
I think a lot of people on this site underestimate the suspicion with which General Counsel’s offices already view LLMs. I bet that there will be a reluctance in US firms to adopt or use DeepSeek even if it is the best open, local model as they are worried about unspecified risks
English
@benspringwater I think some of the confusion is thinking of Lewis as anything _but_ a storyteller. He very openly wants to be Tom Wolfe
vanityfair.com/culture/2015/1…
English
Great little "Theory of Michael Lewis" riff in today's Diff.
Byrne Hobart@ByrneHobart
Today's subscribers-only issue of The Diff: a review of Going Infinite, with firsthand accounts of Lewis' research process. And this:
English
@lessin Challenge with the “Money” approach is that you will attract the worst candidates (terrible but interview well or prestige backgrounds) and you have to be excellent at spotting the difference. With mission/misfits you have an advantage of high signal to noise
English
3Ms To Getting the Real Talent On Your Team (Mission, Misfits, Money)
English
Because, if your e.g. county won the geographic lottery, your landowning class (who have substantial political heft, always) have all just become very substantially richer *if and only if* polity can be convinced to allow fracking.
So they organize to ensure that happens.
English
I learned a bit about fracking today and even relative to my prior expectation holy cow does it sound like sorcery.
As one example of many, how do you think they get information from bottom of the well to top of the well while drilling?
English
@goodside I mean...Not surprising? They tested OOD using Swahili/Filipino/etc while the multilingual models used here were trained on datasets (wikipedia for mbert) with <0.1% representation of those languages.
English
this is wild — kNN using a gzip-based distance metric outperforms BERT and other neural methods for OOD sentence classification
intuition: 2 texts similar if cat-ing one to the other barely increases gzip size
no training, no tuning, no params — this is the entire algorithm:
Luke Gessler@LukeGessler
this paper's nuts. for sentence classification on out-of-domain datasets, all neural (Transformer or not) approaches lose to good old kNN on representations generated by.... gzip aclanthology.org/2023.findings-…
English
@Broncho24 Richard Rodes “Energy” is much more of a general history than most of those above going from wood to nuclear with a lot of context around each transformation.
English
@headphoneDas Had the reverse experience with my kids listening to the lion king. “Wait…is this the gladiator guy”. That guy does everything! Dad lives everyone dies, dad dies everyone else lives, he can score it all
English
@ricburton I'm also of the opinion they should send you a tote bag
English
@parkerconrad @natfriedman @OpenAI +1! From the self-censored examples it seems they might have a secondary model (or separate model head) that detects offense and then produces a stock response - or - a smaller model generates a trained response. Stable diffusion does something similar (but nulls the output)
English
@natfriedman I would really love to understand how @OpenAI trained the model to be inoffensive. Would be fascinating to understand what went into that.
English
These models that have been trained never to offend the most shrill parts of society are going to be like living with an HR person monitoring your every move. The iron prison, in every corner of your life. Hell on earth.
English
@CJHandmer I thought it was pretty fair. He was kind to many of the early projects like Owens Valley (from a water use standpoint at least) and Hoover. He just let it rip on the later projects that were just keeping engineers busy at otherwise negative net value
English
The book is a fascinating historical account but, to me, seems built on shaky premises. An increasingly strident bias as the chapters progress favoring degrowth and rewilding of the west. Not without cost but the BoR and friends really did terraform half a continent.
English
Finally reading "Cadillac Desert" and I've never felt so scolded in my life!
English
@hkarthik Let’s be careful here - you and I can’t both be top 25% in our org! Rather than working too hard, we should pull straws to decide who survives
English
This is my third recession in tech, here's my playbook:
1. Cut spending. Put off vacations, auto, and home purchases. Have 12 months of emergency savings.
2. Put in more effort at work to reach the top 25% of performance.
3. Learn more marketable skills with your extra time.
English
@TerraformIndies Since it’s opex dependent does profitability mean the machines only run during cheap daylight hours?
English
@KipKaehler Each fits on a single flatbed for deployment on skids. Manufactured, deployment involves dropping on the ground, connecting a few plugs and hoses, then walking away.
English
We had a great conversation with a candidate on site at the Terraform Industries castle this morning. One of the key takeaways is that, during the design phase, we have considerable freedom to choose where we put the hard part.
English
@TerraformIndies Whats the size/capacity of each machine? Are we talking manufacturing or construction for each deployment?
English
Building millions of them. It's not enough to make a shiny prototype or proof of concept. We can be profitable with 1000 machines in operation and extremely wealthy with a million. But we won't make a dent in the climate problem with less than 100 million machines.
English
@JayNDonde I agree with your point on unclear goals. We seem to aim to increase the mean education level but I think society is better served if we split resources b/w the 25th percentile (foundational skills for all) and 99th (more Nobel prize winners). Average is a waste for most topics
English
@JayNDonde Awesome thought experiment, Bryan Caplan’s book makes an entertaining case for the “burn it all down” approach.
amazon.com/Case-against-E…
English
@matt_levine @patio11 @tqbf You need to take this opportunity to write our generation’s Barbarians at the Gate. Something about an electric clown car
GIF
English
English
@albrgr The Making of the Atomic Bomb is such an incredible combination of history and science (and the history of the science). Very under rated. If you liked it Dark Sun is an awesome follow up on the science while American Prometheus is a great afterword on the politics.
English
Was scrolling through my kindle library, some random nonfiction I've read over the last few years and would highly recommend:
amazon.com/gp/product/B00…
amazon.com/gp/product/006…
amazon.com/gp/product/145…
amazon.com/gp/product/067…
amazon.com/gp/product/038…
amazon.com/gp/product/159…
English
If you pay $11 billion in taxes you shouldn’t be demonized by elected officials
They should at least send you a free t-shirt or something
English
@kaehler1920 @typesfast Tax meat to lower demand/cost for grains, suspend the tax for Ukrainian livestock so they can convert their wheat->meat and march it across the land border.
English
@typesfast Good point... Are the land adjacent countries dependent on port for delivery? Those ones seem to be the highest % of 🇷🇺 wheat.
Definitely scary for the African countries that are reliant on both.
I'm sure this is in Putin's calculus for how to lift some sanctions.
English
Russian and Ukrainian grain won't make it to market this year. And with fertilizer prices surging, many farmers are scaling back production. What can be done to avoid food shortages around the world? Post your best ideas that don't require acts of congress.
English