tryna make sense of it all
Joined December 2015
- Tweets7.6K
- Following1.3K
- Followers2K
- Likes12.2K
angelo retweeted
Leftists care more for the world than their family/friends because it's easier, not because they're altruistic
Caring for your immediate circle is a tangible daily task that requires effort
Caring for "the world" is an abstract virtue signal that needs zero actual output
angelo retweeted
How many dam collapses happened for hydroelectric to move forward?
How many large passenger jet crashes for commercial aviation to become safe?
How many surgeries killed a patient to have modern medicine?
This is human advancement 101. We forgot this. But it's back.
Lots of very very bad stuff ahead amidst wonder and progress. Leadership is making sure the world hears it straight from the top. As it is. Treating us like adults.
Thank you @sama for saying it straight.
angelo retweeted
This is fucking disgusting and a crime against humanity.
This company is growing real human neurons to use for computing. We don't know how consciousness works! It's bad enough that we don't know what to think of brain-like LLMs. This company says fuck it, remove all doubt, make the computer out of human parts!
If you don't think the neurons they are using now are conscious, do you think this is where it ends? No! Obviously not. Obviously if they can get more performance out of something more brain-like they will. We cannot allow them to use literal human brains as computer products.
And they know they're doing something horrible. They chose to name their product "mischief" and launch with a "happy halloween" nod. The AI companies are attacking our society in places where we don't have laws yet. Our government needs to react fast.
today, we introduce a little mischief to the world.
for four years, we’ve been brewing real brain cells on electrodes to understand how they compute. now, we’ve turned those discoveries into a video model anyone can use.
from 0.5¢ a second, mischief 1 is at least 5x faster than its base and the cheapest image-to-video model on @fal and @OpenRouter. soon to be on @awscloud marketplace.
it’s not a delusion anymore. biocompute is real, commercial and this is just the beginning.
happy halloween.
angelo retweeted
But if it's enjoyable, it doesn't require force, and force doesn't produce learning, and it can cause resentment, and if the forced thing is cherished then it can cause resentment of a cherished thing....
angelo retweeted
“Always, one should think carefully about when to expose an idea to criticism.
If one does it too early, it may cut off lines of thought which although false, would lead to something true if one pursued them seriously.”
— @DavidDeutschOxf
angelo retweeted
In free nations, we investigate crimes to see who is guilty of them.
In communist tyrannies, they investigate men to see what they can be punished for.
angelo retweeted
LearnHarness: an agent harness for learning things.
The goal is strong, usable understanding, with mental effort focused on direct interaction with the material rather than managing the learning process.
The best explanations and the best practice problems depend on what you already understand and how you understand it. So the agent needs a detailed picture of that understanding, continually updated as you learn.
TEACHING
1. Probe
The agent probes your understanding of everything the lesson depends on, building a detailed picture of your mental models and ways of thinking.
2. Plan
The agent reasons out the minimal dependency tree from what you already know to what you want to understand.
3. Teach
Starting from what you already understand, the agent teaches one reasoning step at a time. The aim is for each step, and eventually the goal, to feel obviously true.
At each step:
i) You predict the next move.
ii) The agent explains it.
iii) The agent checks your understanding.
Predictions and checks help surface errors and confusions immediately, so they can be resolved before later steps build on them.
PRACTICE
Following the reasoning and having the right ideas come to mind when needed are different abilities. Practice makes understanding usable by broadening and strengthening cues for recall.
Problems adapt to target what needs strengthening and to stay at an ideal difficulty level: hard enough to require real struggle, but not so hard that you can’t make progress.
When you get stuck, the agent aims to give the smallest possible hints needed to unblock your thinking without replacing it.
SHARED RECORD
Teaching sessions and imported notes are distilled into propositions and reasoning chains. This compact, persistent record is shared between teaching and practice. It renders as a graph.
APP
Subagents research, read PDFs, create diagrams and build interactive demos.
Free, open-source desktop app. Use your own keys and models. Learning records are local plain files. System prompts are visible and editable.
github.com/amosblomqvist/lea…
Using an LLM to teach me matrix multiplication
Realised that matrix multiplication is teaching me matrix multiplication
angelo retweeted
🚨Silikon Vadisi aylardır en zeki sohbet botunu yapmak için milyarlarca dolar yakıyor ama Elon Musk tek bir cümleyle bütün o yazılım devlerine çok ağır bir gol attı:
Yunuslar ve balinalar da çok zeki.
Ellerin yoksa, nesneleri tutamıyorsan, o devasa zekanın ekonomik değeri koca bir sıfıra yaklaşıyor. Piyasa hala yazılımları fiyatlarken asıl yarış fiziksel yapay zeka tarafına kayıyor.
Okyanusun ortasında dünyanın en karmaşık matematik problemini çözen bir balina düşün. Ne bir çivi çakabilir, ne bir mikroçip dökebilir, ne de bir laboratuvarda deney tüpü tutabilir.
Bugünkü dil modellerinin çoğu tam olarak bu durumda: Cam ekranın arkasına hapsolmuş, parmakları olmayan dev birer okyanus beyni.
Zekanın gerçek çarpanı daha hızlı kod yazmasında değil; atomları büküp nesneleri şekillendirmeye başladığı anda açığa çıkıyor.👇
