1,130 karma · joined March 9, 2017
- also a calculator does not offload your critical thinking ability
- fun fact, i have stopped using calculators since the last 3 months or so as an experiment and guess what? I can subtract and add six digit numbers effortlessly now
- also a calculator is not subject to bias which the AI frontier model companies can most certainly push in your direction if they wanted to.
- So when I see people comparing AI with the dawn of calculators, i really sit and wonder how such a comparison even makes sense
their product wouldnt run if they had 0 users
- because it finally clicks and you realize how it works on the inside?
- https://visualrambling.space/neural-network/
- i found one but it doesn't have text content explaining the math, the reason behind gradient descent, the need for a partial derivative etc.
- I would super appreciate if someone has a resource for cnn, rnn, and other types of neural networks like this
Prediction: Microsoft Is Going To Do The Funniest Thing Imaginable this guy called it loooong back
FTFY the headline for you
- https://www.seangoedecke.com/page/2 or
- https://mmapped.blog/posts this or
- https://blog.frankmtaylor.com/archive doesnt work
- https://blog.frankmtaylor.com/archives nope
- https://blog.frankmtaylor.com/posts nada
- https://blog.frankmtaylor.com/all nil
- https://blog.frankmtaylor.com/blog nyet
- do you understand that most of us got 10000+ bookmarks and it would be really really handy if you could make a single page on your website that lists every post every written without the description or the entire article on it?
"4. When the answer feels done, you stop thinking"
how do I run it locally?
- and for the web / javascript / svelte applications?
- suggestions for local OCR for bulk images?
- just like how you ll become good at coding if you stopped using an llm (for most of it atleast)
- i am not saying this to trigger anyone. i am speaking from personal experience
- 6 months ago i stopped using calculators to add, subtract, multiply and my abilities have sharpened slowly and steadily
- initially i had trouble adding / subtracting 3 digit numbers
- i have slowly come to the point where i can do 6 digit numbers now without a calculator and pretty fast
- obviously i wont be able to prove that here because people ll think i am using tools to paste the answer but if were to meet in real life like ever, you would be absolutely blown away by how fast i can math
- was at the bank the other day, the bank manager had a mini heart attack when i calculated the 8.5% interest rate on the 1st year precisely before he could use his calculator
- seriously give it a try. start with 2 digit numbers addition / subtraction for the next entire week and slowly up your game. it ll help you immensely
- not a single post has anything inside here https://flow.byu.edu/posts/
- i am trying to do something along these lines
- What I am looking for is "informationally dense" articles across a blog
- For example, a person writing "Ai is going to do this" and "how I felt at my company when they adopted AI" are pure opinions.
- ON the other hand, a post like "here are 10 ways to loop a directory in bash" is informationally dense
- What sort of techniques / algorithms do you think I could use to narrow it down. I can think of removing stop words from the post, counting the ratio of remaining words to total words (not sure if that means anything), n gram analysis maybe but I am really not an expert at this
- Perhaps someone at HN can shed some light on how to go about identify "information rich" articles on a blog
- Do you think LLMs would do a good job if we were to loop through every post on a blog and ask LLMs to pick non opinion ones
- how do I prove that they are actually not privacy friendly?
- your links to the slides for deeplearning you did here https://sander.ai/2014/05/29/slides-meetup.html are broken
- stier is no joke. it has to explain something in a way that even a 5 yr old understands with animations and visualizations
- the information quality of stier in my criteria has to be top tier, no mistakes no nothing
- here are some of my stier bookmarks
- https://radarlaboratory.com/ teaches you how radar works
- https://nandgame.com/ is game that encourages you to build a computer from scratch like literally from transistor level
- https://shader-learning.com/ exercises to learn how shaders work
- https://timmastny.com/blog/visualizing-cpu-pipelining/ shows you cpu works under the hood
- i have 1000s of such resources, 1000s gotta share em on HN one day
- for whatever reason, even the same sized directory takes much much longer than its corresponding archive version when using this tool
- it is not going to be pretty when they 5x their prices or degrade their models in some way