Godel_unicode

This is pointless word-smithing to get out of being blamed for consequences. These popups exist because of GDPR requiring them, they’re GDPR popups.

GDPR is the prop65 of privacy, and there was no chance of any other result.

Companies would just change the checkbox from “I acknowledge the privacy policy” to “I grant you this license”.

First, for some people front loading their earnings is important so they can enjoy more of life. Second, as another thread goes into, the risk-adjusted picture of startup vs established tech company is quite different.

That’s what the comment you’re replying to is saying, you just used more words. You’re downstream from someone who asserted they are a hard 0 on the caring, which it sounds like you agree is… unlikely.

If you actually don’t care you wouldn’t post because it wouldn’t occur to you to.

Edit: I find that most people who scream “I don’t care what you think” generally mean “I’m not influenced by what you think”. Which is an important difference, and helps explain getting red in the face about someone else’s opinion about which one don’t care.

It would be fascinating to take a sample of people, compare their avatars to their real life appearance, and then have them spend an hour talking to a shrink about why they are so similar/different.

Anecdotally, I know many people in both the very similar and very different camps, and a few who are very similar except for some small thing.

There is no chance that those were the three options at a federal public meeting, primarily because that type of survey is not how public meetings work.

It’s certainly the case that public meetings are not binding on the agency, nor should they be. We already have a process for determining the will of the people and it’s not “turf the problem to the 5-10 people who showed up to this meeting on a random Tuesday”.

Public meetings are one part of a larger process.

I mean, that is how it works, but sometimes agencies overstep their delegated authority and are overruled by the courts. Then they have to go back to the congress and get their authority amended or not.

This is how laws work in the US though, it’s called rule making authority. For example, the clean air act doesn’t specify any of the levels or pollutants, it empowers (or in that case creates) the Department or Agency to make rules within some often quite broad guidelines. That agency will then conduct research, hold public meetings that nobody goes to, and then make rules which are enforced under the law.

Edit: those laws sometimes get amended by congress and sometimes extended or have their enforcement deprioritized by the executive.

> There is no use case for

There are a ton, computers are incredibly useful devices.

If I close the screen I can play music out of those speakers for a pretty long time. And if I can sleep/wake from a Bluetooth keyboard when I’m not using it, I can stretch that into a long weekend.

…or at least I can on Windows where this is possible.

Good luck convincing the HN crowd that the FBI aren’t rubbing elbows with the MSS under everyone’s bed.

Huh? First, the bank can’t possibly price future interest rates into the price of a fixed loan as they’re by definition unknown. Second, we’re talking about the economic model of which is better for consumers. It assumes no refinancing to lower your interest rate down the road, which is a pretty big caveat.

Oh good, more adoption of this exhausting trope of religious texts being called myths. Very inclusive.

That math is my favorite example of economists not living in the real world. It doesn’t account for the ability to refinance.

Not sure why you’re being downvoted, this is a very good answer. Maybe because you left out the implied “and then firewall off that vlan”?

> These are plants that have had genes from another organism inserted into their DNA to give them a new trait, such as disease or drought resistance

Maybe you’ve never experienced good management? Good managers are significantly different in very obvious ways even in the absence of conflict.

A good managers job is not just helping you navigate chaos.

That’s something like a 50% premium over Azure, even before reservation discounts, that’s insane!

Edit: todays pricing looks like about 20% higher, still. How are these prices so different.

This quote has, pound for pound, provided more comfort to terrible leaders than any other. It’s also just not true, good leadership is very obvious.

The sad part is they think that all the reporting and metrics, and the juking of the stats which goes along with it, is making things better. In reality it’s generally keeping things the same but with a pretty coat of paint.

Your conclusion sounds an awful lot like the people in your first sentence are right. Anecdotally, this is what I’ve seen. As people progress on the engineering side and the work gets more demanding, increasingly people bail out and make the transition to management.

You might want to do some research into the words “science” and “pedantry”, since the former is just meticulous application of the latter. Calling a scientist pedantic is a compliment.

That aside, you’ve been arguing that these models understand things and citing these papers as evidence. They are not. They are evidence that the ability of these models to generate text based off of their existing training set can easily be finetuned in a number of ways to add training sets after the initial zero-shot learning.

That’s all these models do, they generate text based upon some training set. If we define understanding as the ability to extrapolate beyond what one has been told, they are expressly not doing that. Your papers explain this quite well.

Edit: to see more concrete examples of this, look into the unfortunately named “hallucination” ability of LLMs. Once you realize that they only know what they were told and are unable to logically extrapolate the point becomes clearer. I hope that helps.

You said “… models end up internally implementing…”. This is incorrect and also importantly different than what you just said. The model and the training using ICL are different things, which it appears you are now beginning to understand.

Precision in speech is critical when discussing complex subjects.

Your second bullet isn’t what the paper says. It says that the model’s accuracy under in-context learning from the meta gradients is similar to manual fine-tuning with back propagation (under some surprising constraints in the name of “fairness”). The language model isn’t implementing anything, you have the directionality wrong.

GPT models are constructed with pretrained gradients which are applicable in a large set of situations. It’s just an optimization technique, albeit a clever one.

Quoting from the paper:

In summary, we explain ICL as a process of meta-optimization: (1) a Transformer-based pre- trained language model serves as a meta-optimizer; (2) it produces meta-gradients according to the demonstration examples through forward computa- tion; (3) through attention, the meta-gradients are applied to the original language model to build an ICL model.

Our ability to prove a theory or not is entirely dependent on our ability to gather evidence for or against. How many other theories can be thought of is totally irrelevant.

Low confidence means inability to prove or disprove. That is it.

Edit: extrapolating further about this is just a Rorschach test.

What it means is that we lack the ability to prove either of those theories, not necessarily that either theory is bad.

“We don’t know” is by definition not a theory though, it’s a lack of a theory.

This has been blind tested time after time, and at best nothing beats out a carefully opened bag that’s folded tightly. Many of the fancy containers are actually worse.

Get someone to do both and blind taste them, he’s right.

I mean, not really? There’s a lot of legacy stuff written in other languages, but malware authors have realized that people are less skeptical of rust and are actively taking advantage of that fact.