I'd be surprised if it was significantly less. A comparable 70 billion parameter model from llama requires about 120GB to store. Supposedly the largest current chatgpt goes up to 170 billion parameters, which would take a couple hundred GB to store. There are ways to tradeoff some accuracy in order to save a bunch of space, but you're not going to get it under tens of GB.
These models really are going through that many Gb of parameters once for every word in the output. GPUs and tensor processors are crazy fast. For comparison, think about how much data a GPU generates for 4k60 video display. Its like 1GB per second. And the recommended memory speed required to generate that image is like 400GB per second. Crazy fast.
Chatgpt is also probably around 50-100GB at most
I use qdirstat a lot to determine what files are eating all my space
The other option is a housing coop. Where you still rent, but it's owned by all the renters collectively.
We were in upstate NY, and got extremely lucky with a hole in the clouds right around the sun at totality.
The red at the bottom was unexpected and very cool to see. It's a solar prominence
This is one reason I'm switching away from pla+ back to normal pla. The esun pla+ really seems to get brittle when held under stress. This is an issue with printed parts as well. I've had parts suddenly crack in half where they were stressed over a few months.
Also it's really annoying when little bits of filament get stuck in your filament guide tube :(
There's definitely software that uses parts of the windows API that games don't touch. And doesn't work properly on Wine. I keep a windows install around just for using an analysis software for some lab equipment that refuses to start in wine.
Things like CAD software are also a struggle, though the latest wine seems to have resolved a number of graphics issues with getting PTC Creo to properly use the nvapi and nvidia graphics drivers through wine.
While wine is amazing, plenty of things don't work with it. Usually you don't need them, but if you do, you do
Mine has a setting to not send more than one notification within X minutes I under settings > notifications > app notifications > some app > minimum time between notification sounds
I think its:
Church
Don't you dare to forget your lord god
Pillar & Ground of the truth
Edit: 1 TIM 3:15 is
https://www.biblegateway.com/verse/en/1%20Timothy%203%3A15
but if I tarry long, that thou mayest know how thou ought to behave thyself in the house of God, which is the church of the living God, the pillar and ground of the truth.
More Edit: and Deuteronomy 6:12 is about not forgetting your lord
How many internet service providers would have to go along before the internet was effectively off? 3? 4?
Firefox PWAs seem to work for me on mobile. To be fair I'm on nightly, but I can see a menu item that says "install" if the webpage has a PWA manifest. I was using voyager with it for a while before they released the play store version.
There are absolutely not 2.8 million active subreddits. I just spent like an hour trying to find data on this. Nobody cites their sources. I used a dump of subreddit statistics from 2018, when there were just over a million subreddits. (Source: https://www.reddit.com/r/ListOfSubreddits/comments/8gzmmv/i_created_a_better_csv_textspreadsheet_list_of/)
There were ~34,000 subreddits with more than a 1000 subscribers. And 100,000 subreddits with more than 125 subscribers.
Looking at https://subredditstats.com/ the top 5000 subreddits make up about 30% (based on an estimated 840,000 posts a day by some reddit user on a subreddit that's currently dark so I can't give a good link) of the daily posts and surely far more than 30% of the daily traffic.
I feel like this really depends on what hardware you have access too. What are you interested in doing?How long are you willing to wait for it to generate, and how good do you want it to be?
You can pull off like 0.5 word per second of one of the mistral models on the CPU with 32GB of RAM. The stabediffusion image models work okay with like 8-16GB of vram.