Well, TinyChat. Still mind-boggling. From the video description:
I built a small language model in Minecraft using no command blocks or datapacks!
The model has 5,087,280 parameters, trained in Python on the TinyChat dataset of basic English conversations. It has an embedding dimension of 240, vocabulary of 1920 tokens, and consists of 6 layers. The context window size is 64 tokens, which is enough for (very) short conversations. Most weights were quantized to 8 bits, although the embedding and LayerNorm weights are stored at 18 and 24 bits respectively. The quantized weights are linked below; they are split into hundreds of files corresponding to the separate sections of ROM in the build.
The build occupies a volume of 1020x260x1656 blocks. Due to its immense size, the Distant Horizons mod was used to capture footage of the whole build; this results in distant redstone components looking strange as they are being rendered at a lower level of detail.
It can produce a response in about 2 hours when the tick rate is increased using MCHPRS (Minecraft High Performance Redstone Server) to about 40,000x speed.
>> One of my fondest memories was buying the book "The Elements of Computing Systems"
>> by Nisan and Schocken, and implementing a 4-bit CPU in Minecraft.
You confused me there, the book doesn't cover Minecraft, you did that yourself after reading the book, got it.
The book is absolutely fantastic, it is the basis for the "From Nand to Tetris" courses:
https://www.nand2tetris.org/
I haven't digested it in full and with a title like that and the boring cover I always have to scramble to find it when I got a few minutes (What is that "Nand to Tetris" book called again?)
> (which I don't think existed at the time anyways)
This surprised me at first because I remember using Python to create Minecraft maps early on, or at least in the beta.
But it seems like redstone was added in the alpha, and the earliest commits of pymclevel (which I think I used) also dates back to the alpha. So there might indeed have been a time window of a few months in which redstone was available but not tooling for creating maps.
I remember starting playing Minecraft Beta, that was around 2011 maybe?
Even if tools to programmatically create maps were available, I wouldn't have known how to program. So this is more about my lack of knowledge of tools of the time.
Funny, I bought that book and studied it because I heard about it in a Minecraft video. Was that you? Nand2tetris helped me land my first programming job.
there were no clickbaits there at all. no command blocks were used at all. if you were so certain, why dont you download the world and try it yourself?
Here is an earlier, longer video on someone explaining how they built a neural network to recognise handwritten digits in Minecraft: https://www.youtube.com/watch?v=DQ0lCm0J3PM
It should answer some of the questions/clear up some of the confusion raised here (e.g. how they get the weights in).
Are these kind of projects actually build manually inside of minecraft block by block or is there some verilog/vhdl to minecraft-level compiler toolchains used?
It’s been awhile since I’ve played Minecraft, but when I built large redstone projects before, I built out each circuit manually and then used mods to copy/paste it within the game.
Oh. I thought this would be some cheesy command block curl to Chat GPT.
But no. The author actually embedded a small LLM in Minecraft using hundreds of millions of blocks, that generates 1 token per 2h at 40,000x speed.
Bravo. I wouldn’t have even thought to try.
4-bit is small enough that you can build it manually, without the use of external tools (which I don't think existed at the time anyways).
Highly recommended for children interesting in computing!
>> by Nisan and Schocken, and implementing a 4-bit CPU in Minecraft.
You confused me there, the book doesn't cover Minecraft, you did that yourself after reading the book, got it.
The book is absolutely fantastic, it is the basis for the "From Nand to Tetris" courses: https://www.nand2tetris.org/
I haven't digested it in full and with a title like that and the boring cover I always have to scramble to find it when I got a few minutes (What is that "Nand to Tetris" book called again?)
This surprised me at first because I remember using Python to create Minecraft maps early on, or at least in the beta.
But it seems like redstone was added in the alpha, and the earliest commits of pymclevel (which I think I used) also dates back to the alpha. So there might indeed have been a time window of a few months in which redstone was available but not tooling for creating maps.
Even if tools to programmatically create maps were available, I wouldn't have known how to program. So this is more about my lack of knowledge of tools of the time.
- Re-implements parts of Minecraft
- Runs 512x512 plots in different threads
- Compiles Redstone applying different kind of optimization passes https://github.com/MCHPR/MCHPRS/blob/master/docs/Redpiler.md
- It had Jit backends before, but seems they have been removed
It should answer some of the questions/clear up some of the confusion raised here (e.g. how they get the weights in).
https://modrinth.com/mod/axiom