The Orthogonality Thesis (by Nick Bostrom) says that intelligence and ultimate goals are independent. Those non-instrumental goals can’t be stupid nor right or wrong. Increasing intelligence will not change the goals…
At first I noticed that for many questions two or three of the answers are obviously wrong. So in many cases the correct answer can be guesses easily. But then I noticed that in 90% of the cases the correct answer is…
We run into this issue when watching Korean movies/dramas. My wife prefers Japanese subtitles and I prefer English/German. I haven’t found a way to enable two subtitles in Firefox (via extensions). So in those cases I…
LLVM (Low Level Virtual Machine) != LLM (Large Language Model)
Why run something in a few CPU cyles on a 40 year old home computer when you can do the same (but worse) on a billion-dollar GPU cluster?
Excuse my ignorance, but you can set up your own Discord server? I thought it‘s a SaaS kind of offering.
When the rocket launches it already moves with earth's rotation at that latitude (and is stationary with respect to the ground). That's in fact the reason why many rocket launch sites are near the equator: Free velocity.
Why is it launching with a horizontal velocity? It initially moves to the right even when pointing up perfectly.
For me the page only shows a title but no content (Chrome, iOS). Or am I missing the joke? (That AI researchers and economists have nothing to say about the topic.)
Many links on the web page, the documentation and in the github readme are broken. Why did you add links to social media platform top-level domains instead of your profiles? The „simulation“ is buggy: The stop and reset…
In some ecosystems like Rust/Cargo the lock file can list a superset of the dependencies that actually make it into the final executable. Crates may conditionally include or exclude dependencies based on enabled…
How do you verify that your verification verifies the right thing? Couldn’t the LLM spit out a nice looking but ultimately useless proof (boiling down to something like 1=1). Also, in my experience software projects are…
I think you are way too optimistic. Even with an antimatter drive and 100% conversion efficiency, such rocket would have a fuel to payload ratio of >1000.
It also uses pentagons in some places because a hexagonal grid can‘t tile a sphere. They made sure that the pentagons are located in water, but this feels like it will add even more edge cases to handle.
I didn't finish reading the article before commenting. Mea culpa! Maybe it would be possible to use Phantasy Star Online's network stack via the streaming exploit. But that would leave the hooking part.
> The game runs on a Nintendo GameCube, a 24-year-old console with a 485 MHz PowerPC processor, 24MB of RAM, and absolutely no internet connectivity. In fact, Nintendo did release an official add-on called the Broadband…
Would that be done in two passes? 1. Render the image shifted by 0.5 pixels in both directions (plus one additional row & column). 2. Apply above formula to each pixel (4 reads, 1 write).
I belive there are many low hanging fruits to improve the speed. In my project the code generation took more than 3 minutes. This is rather annoying if you have just renamed a field in a single “freezed“ class. We could…
Thank you. Which is currently the most capable version running reasonably fast on a 3090 (24GB of VRAM)?
At my project we are forced to use Enterprise GitHub including Actions. Often 80% of the build time consists of up- or downloading intermediate or final build artifacts. This can take more than 40 minutes which is…
Just to avoid potential confusion: WebGPU and WGPU are different things.
Proton, not Photon. ;-) Here is a list with games and their support status: https://www.protondb.com/
mixtral:8x7b-instruct-v0.1-q4_K_M got this correct 5 out of 5 times. Running it locally with ollama on a RTX 3090.
Tried to get it to produce its system prompt and got this: You are Gemini Advanced, a large language model built by Google. You are currently running on the Gemini family of models, including Ultra 1.0, which is…
The paper says that training / reconstruction takes 4 hours. While inference runs at 60fps reconstruction is far from real-time.
The Orthogonality Thesis (by Nick Bostrom) says that intelligence and ultimate goals are independent. Those non-instrumental goals can’t be stupid nor right or wrong. Increasing intelligence will not change the goals…
At first I noticed that for many questions two or three of the answers are obviously wrong. So in many cases the correct answer can be guesses easily. But then I noticed that in 90% of the cases the correct answer is…
We run into this issue when watching Korean movies/dramas. My wife prefers Japanese subtitles and I prefer English/German. I haven’t found a way to enable two subtitles in Firefox (via extensions). So in those cases I…
LLVM (Low Level Virtual Machine) != LLM (Large Language Model)
Why run something in a few CPU cyles on a 40 year old home computer when you can do the same (but worse) on a billion-dollar GPU cluster?
Excuse my ignorance, but you can set up your own Discord server? I thought it‘s a SaaS kind of offering.
When the rocket launches it already moves with earth's rotation at that latitude (and is stationary with respect to the ground). That's in fact the reason why many rocket launch sites are near the equator: Free velocity.
Why is it launching with a horizontal velocity? It initially moves to the right even when pointing up perfectly.
For me the page only shows a title but no content (Chrome, iOS). Or am I missing the joke? (That AI researchers and economists have nothing to say about the topic.)
Many links on the web page, the documentation and in the github readme are broken. Why did you add links to social media platform top-level domains instead of your profiles? The „simulation“ is buggy: The stop and reset…
In some ecosystems like Rust/Cargo the lock file can list a superset of the dependencies that actually make it into the final executable. Crates may conditionally include or exclude dependencies based on enabled…
How do you verify that your verification verifies the right thing? Couldn’t the LLM spit out a nice looking but ultimately useless proof (boiling down to something like 1=1). Also, in my experience software projects are…
I think you are way too optimistic. Even with an antimatter drive and 100% conversion efficiency, such rocket would have a fuel to payload ratio of >1000.
It also uses pentagons in some places because a hexagonal grid can‘t tile a sphere. They made sure that the pentagons are located in water, but this feels like it will add even more edge cases to handle.
I didn't finish reading the article before commenting. Mea culpa! Maybe it would be possible to use Phantasy Star Online's network stack via the streaming exploit. But that would leave the hooking part.
> The game runs on a Nintendo GameCube, a 24-year-old console with a 485 MHz PowerPC processor, 24MB of RAM, and absolutely no internet connectivity. In fact, Nintendo did release an official add-on called the Broadband…
Would that be done in two passes? 1. Render the image shifted by 0.5 pixels in both directions (plus one additional row & column). 2. Apply above formula to each pixel (4 reads, 1 write).
I belive there are many low hanging fruits to improve the speed. In my project the code generation took more than 3 minutes. This is rather annoying if you have just renamed a field in a single “freezed“ class. We could…
Thank you. Which is currently the most capable version running reasonably fast on a 3090 (24GB of VRAM)?
At my project we are forced to use Enterprise GitHub including Actions. Often 80% of the build time consists of up- or downloading intermediate or final build artifacts. This can take more than 40 minutes which is…
Just to avoid potential confusion: WebGPU and WGPU are different things.
Proton, not Photon. ;-) Here is a list with games and their support status: https://www.protondb.com/
mixtral:8x7b-instruct-v0.1-q4_K_M got this correct 5 out of 5 times. Running it locally with ollama on a RTX 3090.
Tried to get it to produce its system prompt and got this: You are Gemini Advanced, a large language model built by Google. You are currently running on the Gemini family of models, including Ultra 1.0, which is…
The paper says that training / reconstruction takes 4 hours. While inference runs at 60fps reconstruction is far from real-time.