ttt3ts

There are also tax benefits to keeping rent high. Essentially you can write off the profit of one property with the "loss" of the other. The amount of "loss" is usually the last rent price. There are caviats of course.

This is getting better. Most providers provide price transparency. The data is dirty and not consistent between providers but definitely a start.

Check out https://turquoise.health/patients a startup scraping and standardizing the data.

Even in those systems if you have more money you can get better and faster care. Something like 60% of Canadians have private insurance. I live near the border and several of my friends have visited the US for to bypass lines.

It is part of the pile. So, ya don't know why this matters. You can easily get it.

Also, dupe, also old https://news.ycombinator.com/item?id=37154633

I find it annoying how social GitHub has become. Posting notifying others when I work on public repos. I don't care for my coworkers to know when I am up at 2am working on open source. I also use separate accounts now.

Vlan is a good idea. Doing that now. Thanks.

It isn't about the browser. It is about search and ads which are not free or open source. There are alternatives but Google's market share is ~92% which is similar level to ie when Microsoft faced it's anti trust case. They pay off mozzila & apple to be default search. Bing has 7% market share and they are the only "real" competition.

The browser is just one more way google maintains it's core search business. E.g. their recently released topics api

An M1 does not have the umph for training :( Just rent or buy a GPU. You can train a 7B model easily on a consumer GPU.

This isn't quite true. Discounted cash price is often LESS than a given insurance pays. You can see this your self if you download price transparency data for your hospital (assuming they bothered to actually comply)

For one they are only in a couple states....

Thinking of insurance in healthcare like normal insurance is not accurate. They are activly involved in setting coding practices and therefore prices. They are entwined and in many ways ARE health care as far as billing is concerned.

(Currently studying for my CPC exam)

Soon as I can sideload another browser with real AdBlock I am switching. Looks like the EU might force apple to allow it :)

Yea, internet is dumb sometimes. Here is some context on the often dumb world of CVE https://www.youtube.com/watch?v=2Mfgjp_aK3I

Have a client with many thousands of csv, json, xml files detailing insurance prices. Fundimentally they all contained the same data but wildly different formats because they were produced by different companies and teams. I used ChatGPT to deduce their format so I could normalize them. Easily underbid their current contractor who was using humans for the work and now I have an easy quarterly billing. :)

TBC, I probably could have optimized tokens but contract was profitable and time critical.

Lots. LLAMA 2 was trained on 4K context windows but can run on arbitrary length just the results become garbage as you go longer.

I refer you to https://blog.gopenai.com/how-to-speed-up-llms-and-use-100k-c... for an "easy" to digest summary

The one you could build for under 2K is last gen hardware.

* Chenbro Rackmount 4U Server Chassis RM42300-F (rack mount case Remove the air filter on 120mm fan. Put two decent 80mm exhaust at rear). * Two used air cooled 3090s. About $650 a piece on ebay. Check slot width and make sure everything will fit on your motherboard. Do a burn in when you get them cause used GPUs can be hit or miss. * 5950x CPU (overkill just had it) * 128GB DDR4 * Motherboard with x570 chipset and dual pcie x16. These will birificate to x8 pcie 4.0 lanes to each GPU. This is enough bandwidth to push GPUs to max IME * 1200W+ ATX power supply. * ebay "u.2 pcie 3.84TB" and adaptor for m.2 NVME slot. (again what I had & it is cheap)

If you're going to really beat the thing I would power limit the 3090s to 320w (from 350w). Perf change is not really notable and keeps temps better.

You can run 70B LLAMA on dual 4090s/3090s with quantization. Going with dual 3090s you can get a system that can run LLAMA 2 70B with 12K context for < $2K.

I built two such a systems after burning that much in a week on ChatGPT.

Which country?

Maybe you should read the text of the first amendment?

Dumb license. Useless.

It is part of the pile. So, ya don't know why this matters. You can easily get it.

Yes but so is the cost TBH. E.g. the bay bridge into SF cost 6.5B.

Thanks! That is exactly what I wanted to know.

You would have to run that in your own data center which is what original poster was comparing to.

Bit confused. Couldn't you just run a Linux VM to do your NAT and only pay normal egress?

$.07 where? That is quite a bit below avg in US. Did you include transmission or only generation?

I wouldn't say it is awesome or anything but plex does everything you mentioned. Supports downloading to phone too which is nice while traveling. I have used it for years since google play went away.

I often respond "shut up or pay me". It works pretty frequently for niche technical software and I land some contract work which I get to open source. The alternative of just fixing their issue never works.

So, if it sounds fun I just fix it. If it sounds boring, pay me

Wow, my big takeaway is don't buy from Oracle. This sounds like an IRS audit.

With normal vendors/suppliers you audit them. With oracle your supplier audits you?!

Yes.

Yea, they don't use anything that heavy. I use https://github.com/searxng as my primary search engine (fork of searx)