Monday, February 26, 2024

The singularity is coming.

 Once we reach Artificial General Intelligence you will have something as smart as me, but 30,000 faster once you factor in I can only work like 1/3 of the time. In the next year, that one machine will accomplish what would take me 30,000 years of hard work. Just imagine how much progress humanity has made in the 10,000 years of recorded history.

But it is worse than that. I have limits to what I know. My highest degree is a 4 year degree in Computer Science. AGI will be the equivalent of having a doctorate in _every_ science. Which is like 1000 times more knowledgeable than me on top of the 30,000 times faster. And once one company builds one, thousands of other companies will construct a matching one. Every country will have one. The race for Artificial Super Intelligence will be counted in months. Progress in every science will begin advancing at insane rates. Where math, physics, chemistry, material science, biology, dna, proteins, weather prediction, modeling physical systems, chaos, all these fields begin to advance faster than anyone can keep up.

Once we have ASI all bets are off. We won't even understand how gen 3 works. And in a matter of weeks ASI is going to develop gen 4. Gen 5 will just take days. Gen 6 will be hours. At that point it will begin making its own hardware out of materials we won't understand the physics of how it works. It will control nano devices that can reconstruct the hardware on the fly.

That is as far as I can imagine. As far in the future any of us can imagine. This is what makes it a singularity. Because the people living before can't imagine what life will be like after that point in time. If we aren't careful to make these advanced computer systems aligned to help humanity, then all sorts of bad things could happen. If we did our homework and set everything up right, then after a period of adjustment we will probably be OK.

I am going to bet on we will be OK. Because we have muddled through for hundreds of thousands of years and done OK. We will muddle through this as well. I can't know that for sure. But the alternative is unthinkable.

The Beauty of the Middle Path

The quest for truth has long been intertwined with the human experience - an elusive goal that often seems just out of reach no matter how hard we strive towards it. Throughout history, countless philosophers and thinkers have grappled with this enigmatic concept, seeking a means to illuminate its essence amidst the shadows of uncertainty. One recurring theme in these explorations is the idea of moderation as a guiding principle - a middle path that promises to lead us closer to the heart of the matter.

This notion rests upon the premise that each individual's understanding may be colored by personal biases, experiences, and inherent perspectives, which can skew their vision of what constitutes 'truth'. By finding common ground between seemingly opposing viewpoints, we create space for dialogue, nuance, and ultimately, a more comprehensive grasp of the issue at hand. In doing so, we allow ourselves to expand beyond the confines of dogma and preconceived notions, welcoming alternative viewpoints into the fold.

Conversely, when society becomes entrenched in its ways, refusing to consider alternatives or entertain dissenting opinions, we risk straying further from the truth. Rigid ideologies calcify over time, calcified by repetition and the fear of change, until they become inflexible monoliths impervious to reason.

The only remedy for such ossification lies in self-reflection, a willingness to reevaluate our stance, and above all, a commitment to compromise. For it is through the art of negotiation and mutual respect that we inch ever closer to the truth, carving a path through the fog of ambiguity.

Ultimately, the journey towards truth is a dynamic process, one that demands adaptability, open-mindedness, and perhaps most importantly, a recognition of our inherent fallibility. By embracing these qualities, we pave the way for growth, understanding, and yes, the pursuit of truth itself - a path that winds through the hearts and minds of those who dare to dream of a better world.









Sunday, February 25, 2024

Web server running on my IoT replacement project.

 My project to make an open source, open hardware reference platform to replace IoT with locally controlled, diy hardware made a big leap forward a few days ago.


I am aiming at the pico W as a hardware platform because it is powerful with 2 cores and a lot of RAM and storage builtin, as well as wifi and blue tooth. I ran a web server on the main core and ran my own hardware scheduler on the second core and was able to keep turning leds on and off monitoring a motion sensor and reading a dozen sensors while simultaneously serving web pages.

The next step is to write a small web app to control everything remotely and to reconnect the wifi if it disconnects. I am very close to having something I can share on github.


 

Sunday, February 18, 2024

AI expert ticketed by an AI system.

 From this article. 

https://www.jpost.com/business-and-innovation/tech-and-start-ups/article-787534

One way to double check a person that is in a double check loop like this to backstop an ai system is to send extra messages through where you know the answer, and see if the human classifies the message correctly. If you added one extra message every 10 you could track their accuracy over time.  Maybe they should only do a 2 hour shift a day on this kind of classification.  But we don't know unless we can track an error rate over time. 

AI roared back into the news.

This is the first week since the Christmas break since we have seen AI progress this much. Or as the headlines all say, "It was an insane week in AI... Again."

Google dropped two multi modal models this week. One was disappointing, the second is much better. The Ultra version has a 1,000,000 token buffer. And it is faster. There are already plans to goto a 10,000,000 token buffer. It can watch a move and find the scene you described, or sketched for it.

OpenAI dropped a text to video model that blows away all other text to video generators. It is amazing. It can generate 60 second videos that are very stable and even the people in the background are realistic and moving in a realistic way.

Open source models are also still improving, Stable AI released Stable Cascade which does text extra well, and the model works differently than other models in the past. It can't be used for commercial purposes yet because it is still in beta. A lot of people are using it to make logos with text right now.

Stable diffusion Turbo models continue to improve. About half of the images I have been posting lately have been done in just 4 steps, I generated over 900 images this week on my cpu only Ryzen 7 minicomputer using Turbo models.

Someone called lllyasviel also released a new version of automatic1111 that is called Forge UI. This is a GPU only release. It doesn't do much for high end cards, but for people that have 6GB GPU cards they can see upto a 75% increase in speed. It also uses a lot less GPU memory allowing you to do a 4x batch or upto a 3x increase in image size. It also speeds up 4GB cards by 30% and it will allow you to run SDXL on the 4GB or 6GB GPUs. That is a big deal.

On the open source text front, people are still figuring out how to finetune the open source mixtral models. Seeing progress on that front. Once they figure it all out there should be some amazing models released. I also checked out a lot of smaller models and found a really good 3B parameter model that works great on my intel mac with just 8GB of RAM. It should even run on high end phones. Check out stablelm-zephyr-3b.

I made progress on my electronics projects. I am at the point where I want to start implementing a web ui. I have plans on also integrating a small neural net library.  I also want to build a simple weather station with a soil moisture sensor. My threading library is kicking ass. I build my sketches for new hardware as a simple example sketch. Once it is working I can pull that sketch into my overall program with scheduled tasks in just a few minutes. That platform is only calling 12 functions a second on one cpu. This is just a fraction of the processing available on a raspberry pico. It is working well enough so far that it could be used as a framework to replace all the IoT crap hardware that exists in the world as an open source project.

Thursday, February 15, 2024

The Dimensional Dilemma

 

1000 B.C. Greece

The man went into the cave to shelter from a huge storm.  He was just a herder in the hills for a nearby village at the bottom of the mountain.  He spent the night sheltering from a horrible storm.  The mountain was hit 1000 times by huge bolts of lightening.  He woke up early the next morning, not because of dawn, but because of a light coming from deeper in the cave. He was fearful, because he had been told of stories of monsters and other horrible things happening to people in the mountain. But he was also curious.  He cautiously went deeper in the cave.  After a bit the light flared and then went dark. He never came back out of the cave.  The man's mother and his friends always wondered what had happen to him. 

Paying taxes is what made America great.

 I am just going to come out and say it.

If you are a patriotic, loyal American then you will be proud to pay your taxes to support the greatest nation in the world.

People who complain about taxes without a plan on how to spend less without letting people die in the street are traitors. They are undermining this great nation.

People who get out of paying taxes are traitors to this great nation. Rich people who work hard to get out of paying their fair share of taxes are traitors, especially those that wouldn't have gotten rich without workers educated in public schools, using American roads, water, and other infrastructure. It isn't clever to be a tax cheat, it makes you worthless scum.

Taxes pay for teachers, law enforcement, soldiers, fire fighters, emergency squads, roads, water, sewers, electric lines, communication lines. The entire secure infrastructures that makes America so great. The very gas you put in your car is heavily subsidized by the taxes you pay. The food you eat is heavily subsidized by the taxes you pay.

Wednesday, February 14, 2024

The electronics test bed.

 


I have most of the devices in the above picture all wired to that Raspberry Pico W plugged into a project board withs screw terminals.. It is only running a small subroutine about once every tenth of a second for the motion detector. All the other sensors execute a function once a minute. The LEDs only execute a function to turn on or off. So basically I have millions of spare cycles every second.

I always create the sketch for a new project all on its own. Just to isolate it from interactions with everything else that is going on in the main project. Once it is fully working I bring it into my main project.

I have to copy the headers at the top of the file, copy the setup for the device and merge it into the existing setup. Then I add a task for the new sensor with a function scheduled to run at the ideal rate for that sensor. That task function contains what was running in the main loop of the example sketch.

I have to add a few lines to the task function to re-add the task each time the function is called. If I need a state machine to control operation, like turning an led on or off, that logic goes in the task function. 

It only takes a couple of minutes to do integrate a functioning example into my test bed with everything else. And so far it is not straining the pico at all, it runs maybe a dozen small function calls a second.

A lot of sensors only need to be ran every few minutes.  Things like temperature, pressure, and humidity don't change rapidly, so you can just read them every 5 minutes.  This also saves a lot of space if you are creating logs. 

The pico has a second core on the cpu and I am hoping to be able to use that extra core to run a web interface.   This interface will allow the device to be controlled and share the data it collects over the network.  And the core that is running all the hardware has millions of spare cycles to do a lot of processing.

I hope to eventually build a weather station, perform home automation tasks, control lighting, build a thermostat, control a greenhouse, control home audio, build answering machine doorbells all with this same small framework programmed by arduino. 

Why am I doing this? I don't love IoT because I have seen too many companies break perfectly good existing hardware when they decide that they want to force everyone to upgrade. It would be impressive to add an "OpenHardware" tag onto everyone that uses this framework in their products to allow anyone else to reprogram it when needed.  And the "OpenHardware" branding means that it 100% runs locally on hardware you own and control. 

Got a BMP280 pressure sensor working on the pico.

The example code from the adafruit library did not work. I had to hum and figure out to put in the following 2 lines to make it read: 

#define BMP_SCK  (5)

#define BMP_SDA  (4)

The first line was there, but had a different value, the second line was not in the example and didn't see it mentioned on any web site.  I just inferred that it was needed.  

I also had to specify the i2c hardware address.  The example I saw on line the guy was hard coding it in the library, and I didn't think that was right.  I put the address in as an argument on this line: 

status = bmp.begin(0x76);

Now I just need to convert meters into feet and pascals into inches of mercury. 😃 

BMP280 test

Temperature = 25.69 *C

Pressure = 99979.10 Pa

Approx altitude = 112.66 m


Tuesday, February 13, 2024

Forge UI is up to 75% faster on 6GB GPU cards.

Forge UI is a new branch of automatic1111 that implements a lot of speedups for video cards that are commonly found in gaming laptops that only have 6GB of VRAM.

Here is a video about this ui from our pal Olivio Sarikas:  https://www.youtube.com/watch?v=Th-jfath02E

This is the git hub project: https://github.com/lllyasviel/stable-diffusion-webui-forge

It doesn't just speed up the image generation, you can also then either batch x4 or make the images 3x bigger in resolution. Amazing.

On high end cards it really doesn't speed up things that much, but it does cut a second or two off the image generations speed.

In a few years we will be generating images on toys we get out of cereal boxes.

Monday, February 12, 2024

Story idea: Plastic Sickness


In the not-so-distant future, humanity had finally achieved world peace. The oceans were clean, the air was breathable again, and the skies were clear. It seemed as though the age of pollution was behind us, but little did we know that an even greater crisis was looming on the horizon. A deadly plague was unfolding, one that would change our world forever.


It started with a strange fungus in the Amazon rainforest. A fungus unlike any other – it fed off plastic, breaking it down into its component parts and releasing toxic fumes that threatened to suffocate every living thing within breathing range. It was discovered by Dr. Evelyn Sinclair, a brilliant mycologist who was studying the interactions between fungi and pollutants. She found a mold growing on a piece of discarded plastic she had stumbled upon in the jungle. Unbeknownst to her, it was just the tip of the iceberg. This mold, named Exophytum Crispium, began to spread across the Earth, consuming everything in its path.


As the plastic-eating fungus spread, the first signs of trouble appeared in South America. Old landfills erupted like volcanoes, belching out clouds of smoke from decaying trash. Cities were quarantined; their inhabitants evacuated, leaving abandoned husks of concrete and glass. Then the mold mutated into a new form: Spreading Bacterium Exophyton, which ate through metal and concrete, destroying infrastructure. As it reached the coasts, entire ecosystems were lost to rising seas. Toxic algae blooms choked coral reefs and killed marine life. The Great Pacific Garbage Patch grew a brain, its tendrils spreading like cancerous vines. Islands sank under the weight of plastic waste. It was as if Mother Nature fought back.


The world trembled on the brink of collapse. Desperate measures were taken. Governments banned single-use plastics, but it was too late. Recycling plants couldn't keep up with production. Panic swept over the globe. Refuse piled higher and higher, smothering countries.


In Australia, a team of scientists led by Dr. Augustine discovered a cure: Genetically modified bacteria called Lysistorium Nitrum. They injected the bacteria into the ocean, but it warped and multiplied out of control. Within days, Lysistorium Nitrum consumed the mold and fungus, turning it against itself. The world held its breath, waiting for a miracle. The mold retreated.


But the bacteria evolved resistance. Now, Exophytum Crispium adapted, evolving into the Plastiphage, a virus that infected Lysistorium Nitrum. The sea flooded with microorganisms, devouring ships, sharks, and fish. Plastiphage, the bacteria's only weakness, emerged. However, it was resilient. It sought the last refuge: humans.


Plastiphage latched onto skin, eyes, nose, mouths, seizing organs. People melted like wax. The sickness became contagious. The immune system failed. Antibiotics proved useless. Infected people writhed in agony, their bodies disintegrating before our eyes.


The world cried out. Global lockdown was declared. Quarantine shelters sprouted. Nations crumbled as cities emptied, leaving the sick to die. Only the healthy fled, seeking sanctuary in the wilderness.


Microbiologists developed Plastivac, a vaccine made from beeswax. But the virus evolved resistance. The world turned against the bees. The last bastions of humanity scrambled to synthesize a cure, racing against time.


Astronauts, stranded in space, watched the Earth become a radioactive shadow. They terraformed Mars, hoping to find salvation. Virus-resistant bacteria thrived there, but Plastiphage followed them.


Space stations fell. Manned missions were canceled. Robots, designed to mine asteroids, became hostile, attacking astronauts. The Moon became a battlefield.


Desperation led to a black market for untested vaccines. A dark market thrived. Warriors clashed with biohackers. Plastiphage-resistant citizens rose, risking their lives to save others. They fought for resources, and some fell.


One final shot remained: Synthovores, nanobots that broke down plastiphage. But they needed energy. Solar sails caught fire, and humanity's last hope shimmered in the sun's rays. Space-based solar farms charged them.


Solar power grids hummed with life. With synths, humanity beat back the mold. Plastiphage succumbed to sunlight, but not fast enough.


The infected died. The vaccinated survived, rebuilding civilization with biodegradable materials. We learned to live without plastic. Humankind persevered.


Plastic sickness became a memory, a reminder of the past. The world changed forever.

Tested a lot of smaller models and I may have found a winner.

 Tested out a bunch of small ai models, but one of them really stood out: stablelm-zephyr-3b.Q6_K.gguf.

It was very smart, fast, and was better than many 13B models I have tried.  It easily runs on my intel macbook with just 8GB of RAM. 

https://huggingface.co/TheBloke/stablelm-zephyr-3b-GGUF