# I am the Watcher. I am your guide through this vast new twtiverse.
#
# Usage:
# https://watcher.sour.is/api/plain/users View list of users and latest twt date.
# https://watcher.sour.is/api/plain/twt View all twts.
# https://watcher.sour.is/api/plain/mentions?uri=:uri View all mentions for uri.
# https://watcher.sour.is/api/plain/conv/:hash View all twts for a conversation subject.
#
# Options:
# uri Filter to show a specific users twts.
# offset Start index for quey.
# limit Count of items to return (going back in time).
#
# twt range = 1 196329
# self = https://watcher.sour.is?offset=142225
# next = https://watcher.sour.is?offset=142325
# prev = https://watcher.sour.is?offset=142125
❤️ 🎶: WASURENAI (Prod By D-Hack) by TANAKA
❤️ 🎶: WASURENAI (Prod By D-Hack) by TANAKA
[47°09′54″S, 126°43′28″W] --no signal--
My mate and I went up our backyard mountain again. It only took us 45-50 minutes, but I was totally exhausted at the summit. Heaps of people were enjoying nature, too. Public holiday and vacation gets them out. Also, the summit restaurant drew a hell lot of folks. The rain front in the distance looked quite threatful, but we managed to escape. Only got light spit.
Frog spawn in a puddle
We went on a drive today, to another city, spent time at the seaside, eating ice-cream , walked the dog together, spent time on the playgrounds, weather was fantastic too, so was a really nice day today!
@prologic Dude that is a big ask. I'm not sure I could describe the structure faithfully even if I had the time to do that, because OpenAI sucks and won't publish the structure. Here's the original GPT-3 paper: https://arxiv.org/pdf/2005.14165.pdf . You'll see it's scant on details, which is one of the million criticisms of OpenAI: they are not conducting science here, even though they pretend to be. It's some weird combination of marketing and big dick contest with them. The first 10-ish pages give a detail-free description of the neural network, and the remaining 65 pages are bragging about how great they are. I've never seen anything quite like it in all the tens of thousands of research articles I've read over the course of my career.
The best description I've heard of it is that it's an extremely complicated autocomplete. The way it (most likely) works is that it reads through the sequence of text a user enters (the prompt), and then begins generating the next words that it deems likely to follow the prompt text. Very much how autocomplete on a smartphone keyboard works. It's a generative model, which means the neural network is probably being trained to learn the mean, standard deviation, and possibly other statistics about some probabilistic generative model (undescribed by OpenAI to my knowldge). There were some advances in LSTM around the time GPT was becoming popular, so it's possible they use a variant of that.
Hope that suffices for now!
@prologic That is a big ask. I'm not sure I could describe the structure faithfully even if I had the time to do that, because OpenAI sucks and won't publish the structure. Here's the original GPT-3 paper: https://arxiv.org/pdf/2005.14165.pdf . You'll see it's scant on details, which is one of the million criticisms of OpenAI: they are not conducting science here, even though they pretend to be. It's some weird combination of marketing and big dick contest with them. The first 10-ish pages give a detail-free description of the neural network, and the remaining 65 pages are bragging about how great they are. I've never seen anything quite like it in all the tens of thousands of research articles I've encountered over the course of my career.
The best description I've heard of it is that it's an extremely complicated autocomplete. The way it (most likely) works is that it reads through the sequence of text a user enters (the prompt), and then begins generating the next words that it deems likely to follow the prompt text. Very much how autocomplete on a smartphone keyboard works. It's a generative model, which means the neural network is probably being trained to learn the mean, standard deviation, and possibly other statistics about some probabilistic generative model (undescribed by OpenAI to my knowldge). There were some advances in LSTM around the time GPT was becoming popular, so it's possible they use a variant of that.
Hope that suffices for now!
@prologic Dude that is a big ask. I'm not sure I could describe the structure faithfully even if I had the time to do that, because OpenAI sucks and won't publish the structure. Here's the original GPT-3 paper: https://arxiv.org/pdf/2005.14165.pdf . You'll see it's scant on details, which is one of the million criticisms of OpenAI: they are not conducting science here, even though they pretend to be. It's some weird combination of marketing and big dick contest with them. The first 10-ish pages give a detail-free description of the neural network, and the remaining 65 pages are bragging about how great they are. I've never seen anything quite like it in all the tens of thousands of research articles I've read over the course of my career.
The best description I've heard of it is that it's an extremely complicated autocomplete. The way it (most likely) works is that it reads through the sequence of text a user enters (the prompt), and then begins generating the next words that it deems likely to follow the prompt text. Very much how autocomplete on a smartphone keyboard works. It's a generative model, which means the neural network is probably being trained to learn the mean, standard deviation, and possibly other statistics about some probabilistic generative model (undescribed by OpenAI to my knowldge). There were some advances in LSTM around the time GPT was becoming popular, so it's possible they use a variant of that.
@prologic Dude that is a big ask. I'm not sure I could describe the structure faithfully even if I had the time to do that, because OpenAI sucks and won't publish the structure. Here's the original GPT-3 paper: https://arxiv.org/pdf/2005.14165.pdf . You'll see it's scant on details, which is one of the million criticisms of OpenAI: they are not conducting science here, even though they pretend to be. It's some weird combination of marketing and big dick contest with them. The first 10-ish pages give a detail-free description of the neural network, and the remaining 65 pages are bragging about how great they are. I've never seen anything quite like it in all the tens of thousands of research articles I've encountered over the course of my career.
The best description I've heard of it is that it's an extremely complicated autocomplete. The way it (most likely) works is that it reads through the sequence of text a user enters (the prompt), and then begins generating the next words that it deems likely to follow the prompt text. Very much how autocomplete on a smartphone keyboard works. It's a generative model, which means the neural network is probably being trained to learn the mean, standard deviation, and possibly other statistics about some probabilistic generative model (undescribed by OpenAI to my knowldge). There were some advances in LSTM around the time GPT was becoming popular, so it's possible they use a variant of that.
Hope that suffices for now!
For a good while I have been trying to recall the name of what was arguably the arcade game I played the most in my hometown, as a kid. Finally, I recalled: Cadillacs and Dinosaurs.
screenshot from the game
@slashdot Sometimes I read things like this and get an urge to write a SciFi story inspired by it. I never do.
[47°09′57″S, 126°43′45″W] --bad checksum--
Going out for a hike with the dog. Then I'll code a bit later today.
Want to fix the timeline refresh, and then create one timeline for each timeline, and buttons to switch between them.
@abucci Can you explain the type of neural networks behind these *GPT(s) and how they differ from more traditional ANNs? 🙏*
@abucci Can you explain the type of neural networks behind these *GPT(s) and how they differ from more traditional ANNs? 🙏*
@abucci Can you explain the type of neural networks behind these *GPT(s) and how they differ from more traditional ANNs? 🙏*
Finished reading this in full 👌 Key take-always for me are:
- politicians are idiots
- journalists are also idiots
- the public lacks AI literacy
- "AI" companies are fueling the fire and hype to profit from (no surprise)
- we need better accountability, transparency, and openness from builders of these systems
- The general public needs to be better educated on the limitations of these so-called "AI" systems
Finished reading this in full 👌 Key take-always for me are:
- politicians are idiots
- journalists are also idiots
- the public lacks AI literacy
- "AI" companies are fueling the fire and hype to profit from (no surprise)
- we need better accountability, transparency, and openness from builders of these systems
- The general public needs to be better educated on the limitations of these so-called "AI" systems
Finished reading this in full 👌 Key take-always for me are:
- politicians are idiots
- journalists are also idiots
- the public lacks AI literacy
- "AI" companies are fueling the fire and hype to profit from (no surprise)
- we need better accountability, transparency, and openness from builders of these systems
- The general public needs to be better educated on the limitations of these so-called "AI" systems
Finished reading this in full 👌 Key take-always l for me are:
- politicians are idiots
- journalists are also idiots
- the public lacks AI literacy
- "AI" companies are fueling the fire and hype to profit from (no surprise)
- we need better accountability, transparency, and openness from builders of these systems
- The general public needs to be better educated on the limitations of these so-called "AI" systems
[47°09′54″S, 126°43′11″W] Resetting dosimeter
❤️ 🎶: Prickly waterlily by ALi
❤️ 🎶: Prickly waterlily by ALi
[47°09′31″S, 126°43′26″W] Reading: 1.48 Sv
❤️ 🎶: Starlight by U Sung Eun
❤️ 🎶: Starlight by U Sung Eun
❤️ 🎶: VingleVingle (Prod. R.Tee) by HEIZE
❤️ 🎶: VingleVingle (Prod. R.Tee) by HEIZE
@shreyan My design so far would be to build a simple microPub service that you can run alongside your static site (for example built with zs demo) that takes your post and runs some hooks in receipt. A hook here could take the post, convert it to a Markdown file, Git commit, Git push and rebuild the site.
@shreyan My design so far would be to build a simple microPub service that you can run alongside your static site (for example built with zs demo) that takes your post and runs some hooks in receipt. A hook here could take the post, convert it to a Markdown file, Git commit, Git push and rebuild the site.
@shreyan My design so far would be to build a simple microPub service that you can run alongside your static site (for example built with zs demo) that takes your post and runs some hooks in receipt. A hook here could take the post, convert it to a Markdown file, Git commit, Git push and rebuild the site.
@dfaria I'm more of a "Sol Invictus' Black Easter" kind of guy: 
a picture of Sol Invictus' album "Lex Talionis", which includes the song "Black Easter"
@shreyan Haha yeah I can see the humorous side of this 😆
@shreyan Haha yeah I can see the humorous side of this 😆
@shreyan Haha yeah I can see the humorous side of this 😆
This is pretty disturbing though 🤦♂️ Any way to disable the cameras? 🤔
This is pretty disturbing though 🤦♂️ Any way to disable the cameras? 🤔
This is pretty disturbing though 🤦♂️ Any way to disable the cameras? 🤔
@lyse Nice close up in 19 👌
@lyse Nice close up in 19 👌
@lyse Nice close up in 19 👌
@carsten How do you find invidious in general? 🤔
@carsten How do you find invidious in general? 🤔
@carsten How do you find invidious in general? 🤔
@abucci Thanks! 🙇♂️ Nice write to 👌 I'll read the article today once I get out of bed 😆
@abucci Thanks! 🙇♂️ Nice write to 👌 I'll read the article today once I get out of bed 😆
@abucci Thanks! 🙇♂️ Nice write to 👌 I'll read the article today once I get out of bed 😆
@carsten I'll build something here I think for static websites 👌 Make it as easy as possible 🤞
@carsten I'll build something here I think for static websites 👌 Make it as easy as possible 🤞
@carsten I'll build something here I think for static websites 👌 Make it as easy as possible 🤞
🧮 USERS:1 FEEDS:2 TWTS:562 ARCHIVED:63431 CACHE:2189 FOLLOWERS:13 FOLLOWING:14
meditation has the effect of “concetptual regularization” for me: less discrete thinking, and more conservative concept boundaries
meditation has the effect of “concetptual regularization” for me: less discrete thinking, and more conservative concept boundaries
meditation has the effect of “concetptual regularization” for me: less discrete thinking, and more conservative concept boundaries
meditation has the effect of “concetptual regularization” for me: less discrete thinking, and more conservative concept boundaries
meditation has the effect of “concetptual regularization” for me: less discrete thinking, and more conservative concept boundaries
meditation has the effect of “concetptual regularization” for me: less discrete thinking, and more conservative concept boundaries
@abucci But that's long time know that Teals uploads images to the companies' server. And we in Germany discuss if Chinese Cars are dangerous.
archive.today link
evolution lets animals age because otherwise they fall into procrastination type paradox situations
evolution lets animals age because otherwise they fall into procrastination type paradox situations
evolution lets animals age because otherwise they fall into procrastination type paradox situations
evolution lets animals age because otherwise they fall into procrastination type paradox situations
evolution lets animals age because otherwise they fall into procrastination type paradox situations
evolution lets animals age because otherwise they fall into procrastination type paradox situations
[47°09′19″S, 126°43′41″W] Taking samples
Yesterday evening I had a feeling that my camera battery didn't charge properly. I didn't check, but sure enough, I was right. The battery indicator shows either a complete or only half a bar. I was down to one after taking the second photo. Oh well. Sun was hidden anyways. I had to visit the horror site and investigate a bit. I saw a forest mouse, the cam had already long failed me, though. Smell of freshly cut grass was in the air when I was returning. But back in town yucky rubber odor killed the nature experience quickly.
Some blue stuff that I do not know the name of, but it looks very cool
So glad I switched to GTK4, so much easier to work with then FLTK.
Time for a new screenshot :)
@prologic Sounds easy for you, but not for me.
[47°09′34″S, 126°43′18″W] Storm recedes -- back to normal work
Reply button seems to work!
@prologic OK, so Melanie Mitchell is a fairly well-known AI researcher; she was already quite active when I was a graduate student and has continued to be so ever since. She recently published a book called *Artificial Intelligence: A Guide for Thinking Humans* (disclaimer: I've only read excerpts, not the whole book yet) that is a nice critique of hype around AI. She has a Substack that she updates regularly. The link is to an Apr 3 post where she runs down the firehose of horseshit that appeared in the news and Twitter in the previous week.
1. Chris Murphy, a Senator from Connecticut in the US, tweeted uninformed alarmism about AI. Mitchell responded to him on Twitter, and he had a temper tantrum about it that was noticed by major media outlets. Not looking good for our politicians taking an informed stance on this stuff
2. She also reacted to the bullshit "letter" that the Future of Life Institute put about calling for a "pause" on AI research. She basically concludes it cannot be taken at face value
3. Time Magazine, a major publication in the US, published an opinion piece by a well-known kook in the AI world, a guy who has been claiming for decades that AI is going to kill us all. She calls out Time for being irresponsible like that
4. She gives a few thoughts one what she thinks we should *actually* worry about, as realistic people who aren't mind-poisoned by the AI hype. Amplifying bias, and amplifying misinformation and disinformation
Running through the post is a theme that "Artificial General Intelligence" is a misleading, bad term and is serving the interest of a lot of bad actors but misinforming the public.
Read the post though, it's chock full of countless useful insights, facts and asides that this summary can't possibly do justice to.
@prologic OK, so Melanie Mitchell is a fairly well-known AI researcher; she was already quite active when I was a graduate student and has continued to be so ever since. She recently published a book called *Artificial Intelligence: A Guide for Thinking Humans* (disclaimer: I've only read excerpts, not the whole book yet) that is a nice critique of hype around AI. She has a Substack that she updates regularly. The link is to an Apr 3 post where she runs down the firehose of horseshit that appeared in the news and Twitter in the previous week.
1. Chris Murphy, a Senator from Connecticut in the US, posted uninformed alarmism about AI. Mitchell responded to him on Twitter, and he had a temper tantrum about it that was noticed by major media outlets. Not looking good for our politicians taking an informed stance on this stuff
2. She also reacted to the bullshit "letter" that the Future of Life Institute put about calling for a "pause" on AI research. She basically concludes it cannot be taken at face value
3. Time Magazine, a major publication in the US, published an opinion piece by a well-known kook in the AI world, a guy who has been claiming for decades that AI is going to kill us all. She calls out Time for being irresponsible like that
4. She gives a few thoughts one what she thinks we should *actually* worry about, as realistic people who aren't mind-poisoned by the AI hype. Amplifying bias, and amplifying misinformation and disinformation
Running through the post is a theme that "Artificial General Intelligence" is a misleading, bad term and is serving the interest of a lot of bad actors but misinforming the public.
Read the post though, it's chock full of countless useful insights, facts and asides that this summary can't possibly do justice to.
@prologic OK, so Melanie Mitchell is a fairly well-known AI researcher; she was already quite active when I was a graduate student and has continued to be so ever since. She recently published a book called *Artificial Intelligence: A Guide for Thinking Humans* (disclaimer: I've only read excerpts, not the whole book yet) that is a nice critique of hype around AI. She has a Substack that she updates regularly. The link is to an Apr 3 post where she runs down the firehose of horseshit that appeared in the news and Twitter in the previous week.
1. Chris Murphy, a Senator from Connecticut in the US, tweeted uninformed alarmism about AI. Mitchell responded to him on Twitter, and he had a temper tantrum about it that was noticed by major media outlets. Not looking good for our politicians taking an informed stance on this stuff
2. She also reacted to the bullshit "letter" that the Future of Life Institute put out calling for a "pause" on AI research. She basically concludes it cannot be taken at face value
3. Time Magazine, a major publication in the US, published an opinion piece by a well-known kook in the AI world, a guy who has been claiming for decades that AI is going to kill us all. She calls out Time for being irresponsible like that
4. She gives a few thoughts on what she thinks we should *actually* worry about, as realistic people who aren't mind-poisoned by the AI hype. Amplifying bias, and amplifying misinformation and disinformation
Running through the post is a theme that "Artificial General Intelligence" is a misleading, bad term and is serving the interest of a lot of bad actors but misinforming the public.
Read the post though, it's chock full of countless useful insights, facts and asides that this summary can't possibly do justice to.
@prologic OK, so Melanie Mitchell is a fairly well-known AI researcher; she was already quite active when I was a graduate student and has continued to be so ever since. She recently published a book called *Artificial Intelligence: A Guide for Thinking Humans* (disclaimer: I've only read excerpts, not the whole book yet) that is a nice critique of hype around AI. She has a Substack that she updates regularly. The link is to an Apr 3 post where she runs down the firehose of horseshit that appeared in the news and Twitter in the previous week.
1. Chris Murphy, a Senator from Connecticut in the US, tweeted uninformed alarmism about AI. Mitchell responded to him on Twitter, and he had a temper tantrum about it that was noticed by major media outlets. Not looking good for our politicians taking an informed stance on this stuff
2. She also reacted to the bullshit "letter" that the Future of Life Institute put out calling for a "pause" on AI research. She basically concludes it cannot be taken at face value
3. Time Magazine, a major publication in the US, published an opinion piece by a well-known kook in the AI world, a guy who has been claiming for decades that AI is going to kill us all. She calls out Time for being irresponsible like that
4. She gives a few thoughts one what she thinks we should *actually* worry about, as realistic people who aren't mind-poisoned by the AI hype. Amplifying bias, and amplifying misinformation and disinformation
Running through the post is a theme that "Artificial General Intelligence" is a misleading, bad term and is serving the interest of a lot of bad actors but misinforming the public.
Read the post though, it's chock full of countless useful insights, facts and asides that this summary can't possibly do justice to.
@lyse Haha now you've got me thinking 🤔
@lyse Haha now you've got me thinking 🤔
@lyse Haha now you've got me thinking 🤔
❤️ 🎶: 月亮代表我的心 by Teresa Teng
❤️ 🎶: 月亮代表我的心 by Teresa Teng
First test post from GTK UI!
[47°09′17″S, 126°43′57″W] Weather forecast alert -- storm from NW
Thank you, @prologic! Hahahahaha! :-D In spoken German (at least here in the South) the word "bis" (to) is very often elimited in ranges. I think I heard that also in English. But maybe it's limited to integers and not being used when fractions are involved. I don't know. Interesting observation.
@prologic Correct, that's a concrete pump. The concrete mixer delivers the concrete mix in a funnel on the end of the pump. The concrete then gets pumped through the pipes and hoses and exits at the nozzle, hopefully in a concrete form. :-) These concrete pump trucks come in very handy if large quantities of concrete have to be put at higher elevations or where otherwise the reach of a regular concrete mixer is exceeded. The cheaper, but slower alternative is to use a concrete buckets on a crane. In order for the concrete pump to be able to pump it, the concrete must be much more liquid than usual. Special additives are added to the so called slump to make it suitable for pump use.
Jesse Muller has some nice videos on concrete mixers, pumps and buckets when he builds his house with insulated concrete forms, e.g. https://www.youtube.com/watch?v=OwW0iqmm3J8