fl7305

24 ms·

331.
by fl7305 3y ago
In Sweden, the union can blockade a company that they are "negotiating" with. That means that all other unions will refuse to work with that company, and this goes for both suppliers and customers of the blockaded company. No more
332.
by fl7305 3y ago
Correct. The courts can annul a baseless termination. But the employer can then still refuse to take the employee back. In that case the employer owes the employee 6-32 months pay in damages, and the employment contract becomes legally term
333.
by fl7305 3y ago
> Is it something in the metal itself? Yes. I never understood it myself, until the first time I held a heavy gold necklace. The feeling is hard to describe.
334.
by fl7305 3y ago
"The Bürgeramt also wants to know that you live on the second floor on the right." Meanwhile in Sweden, the government instituted a nationwide "Dwelling Units Register", where each apartment etc is uniquely numbered. The
335.
by fl7305 3y ago
It might be common, but the German bureaucracy is a lot worse than what Germany's reputation of structure and order would lead foreigners to believe. And while it might be common around the world, Germany is playing in a lower league t
336.
by fl7305 3y ago
> Is there any technology you can name that will lead to this notion or an AI improving under its own cognizance? Add a supervising GPT-4 instance that decides which data to LoRA-train on?
337.
by fl7305 3y ago
> I am waiting for someone publishing the theoretical limits of these "AI" systems. > To know the limits turns their application from hype into engineering. It would be helpful to know how the models actually work under the
338.
by fl7305 3y ago
I would imagine that it would do better if it was told to build it up step by step, and critique its own work as it goes along.
339.
by fl7305 3y ago
> people who can't figure out why their code can't produce the same result twice It is well known that the Nvidia parallel processing optimizations cause non-deterministic results. It's easy to get deterministic results as
340.
by fl7305 3y ago
> Nothing prevents you from processing an array in order in parallel. 100% correct if you remove processing time from the equation. In reality, Nvidia Cuda calculations run much faster if you let it schedule the order of floating points
341.
by fl7305 3y ago
I watch a lot of Youtube videos for DIY stuff like car repair and home improvement. The like/dislike ratio used to be a very good way to quickly see if the person who made it knew what they were talking about. Now I instead have to spe
342.
by fl7305 3y ago
A submarine is like a ninja. It tries to sneak around undetected. Using active sonar is like placing a siren on your head.
343.
by fl7305 3y ago
The Romans are said to have used war pigs against elephants. They set the pigs on fire and sent them towards the elephants, who panicked and trampled the people around them.
344.
by fl7305 3y ago
If GPT-4 has 10 terabytes (10^13) of stored memories, how long are the sentences it has stored if there are 100 (10^2) different words to choose from? Wouldn't that be a table of sentences less than 7 words long? 100^7 = (10^2)^7 = 10^
345.
by fl7305 3y ago
I agree. Another way to come to this conclusion is that the GPT-4 model is on the order of 1000B parameters, or on the order of a terabyte of data. That's not a lot of different sentences when you consider the combinatorical explosion.
346.
by fl7305 3y ago
I believe GPT-4 is able to perform some level of logical reasoning. Simpler models look very good on the surface, but there are some examples where this starts to break own. Example: "Is it legal for a man to marry his widow's sis
347.
by fl7305 3y ago
> current GPT style AI does not “think” in the way we think of humans thinking On the contrary, isn't "step by step" reasoning exactly what we tell students to do to become better at solving complicated problems?
348.
by fl7305 4y ago
Ehh, for VW at least some jobs are a lot easier with VW tools.
349.
by fl7305 4y ago
I bought the course a few years back, and am really happy with the price/value in its current state.
350.
by fl7305 4y ago
Good tip. If you really want to work on a car but have a choice in which one, do some research. Cars vary a lot on how easy they are to work on.
351.
by fl7305 4y ago
I never wrote that I was "confident" about it. I fly paragliders for fun, but that doesn't compare at all in excitement to starting an engine where I just changed the timing belt. I watched more than a few hours of video to d
352.
by fl7305 4y ago
That might help, but I've also seen people who grew up in the most sedate middle class families have that kind of natural composure.
353.
by fl7305 4y ago
I went from zero knowledge to changing my timing belt with the aid of the mfg manuals and YouTube videos. A general series like this really helps putting individual YT videos into context. A YT video on changing the timing belt might show m
354.
by fl7305 4y ago
If you want your LLM to predict the name of the murderer in a mystery novel, it can only do that reliably by understanding the text.
355.
by fl7305 4y ago
So your claim is that GPT-4 is a stochastic parrot? Meanwhile, the creators at OpenAI claims that it is not. How do we test your claim? What are the logical reasoning questions we can ask GPT-4 that it shouldn't be able to answer if it
356.
by fl7305 4y ago
> none of them thought to use a nearby heavy steel trolley as a battering ram It has been my experience that most people will freeze up more or less in an emergency. Some people will become completely catatonic, while others "just&q
357.
by fl7305 4y ago
What do "debuggability" or "copyrights" have to do with: Is "GPT-4 is just a parrot" a falsifiable claim? If so, what does the test look like that would falsify it? Can you think of questions to test GPT-4s abi
358.
by fl7305 4y ago
Sure. But my point was that it is not an inherent feature in LLMs that they are frozen in time. Fine-tuning the entire model is very expensive. But fine-tuning a tiny parallell piece using LoRA is cheap both in CPU cycles and storage. OpenA
359.
by fl7305 4y ago
> Running inference does not change the trained weights. That's true today. I don't know how many days (or hours) away we are from GPT-4 running a LoRA-pass to update its weights after each round though?
360.
by fl7305 4y ago
Thanks, I appreciate it! > There have been media headlines about the potential for modern AIs to turn evil and destroy the human race sci-fi movie style. Yeah, the public debate isn't very balanced in either direction.

More ›