svnt

9 ms·

151.
by svnt 7mo ago
That was the limit of just one experimental approach that was peer reviewed and published in a major journal. As you can see there are many experiments validating the limit and none invalidating it. The reality is that the Landauer limit is
152.
by svnt 7mo ago
This was published right before people started experimentally validating the Landauer limit. I am not sure why it hasn’t been taken down at some point as the evidence has accumulated: 2012 — Bérut et al. (Nature) — They used a single colloi
153.
by svnt 7mo ago
For people who want to ask a model for an app, or a website, or something at a level of “hey you make apps right, I have had this idea for years…” the experience is akin to a slot machine — sometimes they get what they imagined their descri
154.
by svnt 7mo ago
Only if the model is actually a human or equivalent, otherwise we don’t know what it is.
155.
by svnt 7mo ago
That’s because it is literally just a feedback loop?
156.
by svnt 7mo ago
The readme opens with this: > I have an RTX 5070 with 12 GB VRAM and I wanted to run glm-4.7-flash:q8_0, which is a 31.8 GB model. The standard options are: > Offload layers to CPU — works, but drops token/s by 5–10× because CPU
157.
by svnt 7mo ago
It’s funny and also disadvantages everyone who can’t trade 24/7. Win/win?
158.
by svnt 7mo ago
I’m confused by this discussion. It seems like you said the biases were structural because we know who reports early and that is why the early numbers are always revised down. Structural implies known in advance. It also seems like you said
159.
by svnt 7mo ago
You use a trailing stop loss. You get closed out 15% down from the top, not 15% down from purchase. The alternative in a 24 hour market is worse — the news of a real event hits and by the time you wake up and respond you’re down 50% or more
160.
by svnt 7mo ago
No, they stop hunt their way to depressed prices where they then buy anticipating the recovery while you closed out your “safe” retirement positions at -15%.
161.
by svnt 7mo ago
What you wrote suggests the data will bias predictably, which matches my understanding. Reporting biased data as the default because the bias compensation is already built into the audience seems like a weak argument for not improving. They
162.
by svnt 7mo ago
I’d counter that if we were doing a good job gathering data that these structural biases could be compensated for with more conservative initial numbers. At some point a lack of decision to take compensating action becomes faking the number
163.
by svnt 7mo ago
That is a reasonable position, however the assumption that it is the administration that is gaming them vs other motivated parties is open for discussion.
164.
by svnt 7mo ago
There is an overlay of smeared poop on one of the license files… is that something you are seeing on typical tech company landing pages? The company is literally named “bad/evil.”
165.
by svnt 7mo ago
The pathos paradox: the more times a person introduces the word pathos in casual conversation the less likely they are to recognize humor/satire.
166.
by svnt 7mo ago
The performance gap between Apple’s flash and a typical aftermarket NVMe drive in a Windows laptop is more attributable to controller design and integration than to trace length.
167.
by svnt 8mo ago
Yeah but why is working for Netflix even sexy? At the end of the day it is almost a telecom.
168.
by svnt 8mo ago
It is interesting that both "customer support" comments here suggesting "you're just using it wrong" are from very recent accounts with very little karma.
169.
by svnt 8mo ago
First you take a 50 person org. Then (for scale) you hire highly motivated performers who, because they came up in big orgs, are used to using 50 people for three years to do a project six people can do in three to six months. Then you crea
170.
by svnt 8mo ago
It is the benchmark error rate, not the benchmark success %, that we actually trip up on. Going from 85% to 90% is possibly 1/3 fewer errors or even higher, depending on the distribution of work you’re doing.
171.
by svnt 8mo ago
I get plenty of creepy engagement hacking with claude’s chat too now. Before memory it was better.
172.
by svnt 8mo ago
It is just Google’s business model, and why OpenAI has to do ads better faster. But you can bet there was more economic foresight going on at Google than OpenAI.
173.
by svnt 8mo ago
How would you navigate to it if you were not?
174.
by svnt 8mo ago
My favorite part is that high pressure hydrogen leaks can auto-ignite in air. http://www.icders.org/ICDERS2007/abstracts/ICDERS2007-0255.p...
175.
by svnt 8mo ago
These people often never knew in the first place.
176.
by svnt 8mo ago
There is value in taking a product to market and hardening it, and no one wants to invest in something that requires headcount for cost-savings. They want upside. But if it doesn't require headcount and/or unlocks functions they h
177.
by svnt 8mo ago
The challenge to this is that so much of the difficulty in getting people to switch products is trust, and a couple of people running saas with claude code has no differentiation and no durability. I think it will be a little different: bla
178.
by svnt 8mo ago
Maybe I am misunderstanding something but: 1) This is basically the intention of several recent MoE models: keep particular generally useful experts hot in VRAM. 2) Unless you can swap layers in faster than you consume them there is no poin
179.
by svnt 8mo ago
Yes but you concede it is a schematic.
180.
by svnt 8mo ago
Maybe the definitions are shifting, but in my experience “on point” is typically an endorsement in the area of “really/precisely good” — so I think what you mean is “on topic” or similar. Pedantry ftw.

More ›