pera

What do you mean by "processing speed"? I think you are interpreting this experiment as if our brains worked like digital computers doing signal processing in some buffer in the visual cortex :p

The term "extermination camp" is more precise for that case.

That article doesn't seem right, or at least some academics seem against the use of "internment camp" in this case:

https://www.npr.org/sections/publiceditor/2012/02/10/1466917...

Why is the BBC referring US concentration camps as "internment camps"?

Another one six months ago: University of Edinburgh orders investigation into Oracle finance disaster

https://news.ycombinator.com/item?id=33739657

There is something wrong with the URL: https://css.ethz.ch/en/services/digital-library/articles/art...

Maybe they fired too many people this time? https://news.ycombinator.com/item?id=35334705

Yeah the way he conflates "crypto" to refer both cryptography and cryptocurrency, and the rhetoric itself is quite odd: https://urbanists.social/@sam/110340265606422596

It's unfortunate because there are some valid points in his criticism.

Already posted 4 days ago: https://news.ycombinator.com/item?id=35835787

Ron Wyden (D-Ore.)

This isn't the first time though: https://techcrunch.com/2016/04/13/burr-feinstein-encryption-...

Hi there, I have two question:

1 - Why did you choose Markdown? It seems an odd choice for training a model like this.

2 - Have you tried to train only one single PL and then benchmark it against this more general version?

their pytorch_model.bin is 10.4GB

Eh so they are making fun of their own customers because of their opinions regarding the "oaty" copyright case? Very weird PR campaign.

> convince gullible investors that there's money to be made with foundational models. There's not.

This is a ridiculously myopic statement. Foundation models are an extremely powerful technological advancement and they will shake the global economy as very few things did in human history. It's hard to imagine how this is not obvious to everyone right now, specially here in this forum.

Valve is in Bellevue and Nintendo is next to the Microsoft campus in the outskirts of Redmond. There is nothing in North Bend (except Twin Peaks fans and hikers).

Then there is no reason for the PD to exist anymore, it's a total waste of taxpayer's money.

Yeah that's what I use as well, but I only use plastic for storing cold things like salad.

Well I for one thank unions for the weekends and 40-hour work week (in addition to many other great things we now give for granted).

And 2019 was the year when they announced the cryptocurrency and the "metaverse" projects...

Yeah sounds like textbook bank run or cryptocurrency exchange collapse, it's bizarre that this kind of things keep happening every few years.

Looks good but I wouldn't run commands without reviewing them first. It would be better if this was integrated to a shell, just as other forms of completion.

Maybe it doesn't have a measurable effect but the other day there was an economist on Bloomberg saying that "we need a recession" that it would be "good thing", I was a bit surprised to hear this sentiment being expressed so explicitly.

Edit: I can't find the video now but I found another recent video expressing a similar idea: https://www.youtube.com/watch?v=63aDL326IFA

I don't understand why your conclusion is that "the model must be thinking beyond the next token": the model doesn't need to do that to generate a well-formed sentence because it's not constrained by the size of the sentence.

The answer is quite simple: unions cannot be right even if they are, empirically. The solution to this dissonance is to ignore them.

That was Rackspace: https://techcrunch.com/2023/01/06/rackspace-ransomware-data-...

Let's say you work in the station and your desk is right next to one of these uranium oxide glazed tiles:

- You work about 2k hours per year in the orange office

- That would then be about 200 mSv per year

- "Exposure to 100 mSv a year is the lowest level at which any increase in cancer risk is clearly evident" (https://www.reuters.com/article/us-japan-quake-radiation-idU...)

I guess then that maybe we could (barely) observe a higher incidence of cancer for those who spend a lot of time next to the walls?

Phones are awful if you are doing exercises though.

I remember the first time I read "OpenAI" I thought it was a new project of the Khronos Group consortium

Well ADS-B Exchange just got suspended (https://twitter.com/ADSBexchange) so it seems that publishing public radio data is now against their TOS.