chayleaf

joined 2 years ago
[–] chayleaf@lemmy.ml 2 points 2 months ago

I use sway on my phone, had to add a secondary menu bar with a few keys for stuff like opening rofi, but it works perfectly fine otherwise

[–] chayleaf@lemmy.ml 2 points 5 months ago* (last edited 5 months ago)

different neural network types excel at different tasks - image recognition was invented way before LLMs, not only for lack of processing power, but also because the previous architectures didn't work with languages. New architectures don't appear out of thin air, they are created with a rough idea of what we could need to make the network do a certain task (e.g. NLP) better. Even tokenization isn't blind codepoint separation but is based on an analysis of languages. But yes, natural languages aren't "parsed" for neural networks, they don't even have a formal grammar.

[–] chayleaf@lemmy.ml 1 points 5 months ago (2 children)

i'm not talking about knowing about how humans perceive/learn languages, i'm talking about language structure. Perhaps it's wrong to call it "how languages work"

[–] chayleaf@lemmy.ml 6 points 5 months ago (4 children)

While I agree that LLMs can achieve human-tier efficiency at most tasks eventually (some architectural changes will be necessary, but the core approach seems sound), it's wrong to say it's modeled after the human brain. We have no idea how brains work as they're super complex, we're building artificial neural networks from the ground up. AI uses centuries' worth of math, but with our current maths knowledge the code isn't too complicated. Human brains aren't like that, they can't be summed up in a few lines of code because DNA is a huge mess that contains so much more than just "learning", so many inactive or redundant bits and pieces. We're building LLMs with knowledge of how languages work, not how brains work.

[–] chayleaf@lemmy.ml 3 points 5 months ago* (last edited 5 months ago)

this kind of software is mostly used for tech support, so your option is too hard to setup

[–] chayleaf@lemmy.ml 1 points 6 months ago

by default, your content is all rights reserved, the most restrictive license possible. AI trains on "all rights reserved" content all the time. You really think adding a CC-BY-NC is gonna do anything?

[–] chayleaf@lemmy.ml 2 points 6 months ago (2 children)

homophones are common in Chinese and Japanese because there's only so many potential readings of a hieroglyph, but each one has a different meaning

[–] chayleaf@lemmy.ml 32 points 7 months ago (2 children)

because killing birds isn't a task of the kernel, it's the task of a userspace utility part of the coreutils

[–] chayleaf@lemmy.ml 1 points 7 months ago

over a century ago Lenin has defined imperialism as capitalism in decay, monopoly capitalism, capitalism that has outgrown competition, that has stopped playing a progressive role in history and became solely a force of reaction, and since then not much has changed

[–] chayleaf@lemmy.ml 10 points 8 months ago

they are, the titles just got changed

[–] chayleaf@lemmy.ml 1 points 8 months ago

it's probably caused by fast shutdown

 

My biggest blog post yet, and it's about running (almost) vanilla NixOS on a (formerly) Android phone! This was 50% fun and 50% exhausting... you solve one issue and another one crops up right away... it was certainly an interesting educational experience.

I'm not explaining any basic technical concepts here, as I'm not a complete noob in phone ROMs and Linux.

Ask me any questions if you have them!

 

This is a lightweight alternative for Goldberg for the single purpose of unlocking DLCs. Just rename the game's steam_api.dll to steam_api.orig.dll, download steam_api.dll from releases and put it in place of the original steam_api.dll, the game will keep interacting with Steam as usual but it will consider all DLCs installed. Of course, you also need to download the actual DLC files from somewhere.

I've actually only tested it on Linux, so I'm curious to know if it works fine on Windows and MacOS.

view more: next ›