Pretty bold to assume I’m not a hunter gatherer

WSL is the best thing that’s ever happened to windows

Hated by everyone he meets. Raises son in hope of creating a potentional liver donor for the decades of alcohol abuse he’s about to experience.

Dad, trying to find jokes to bring light into his miserable live.

Love it. Don’t know why.

In my case, I have to verify online purchases on my bank’s app. Which makes online banking impossible without an android or apple phone.

Color suggests mint

In Belgium, it’s above 60% if you take into account RSZ (mandated health insurance).

2019 data (1) comparing federal income tax.

In Belgium, you’d be at 45% to 50% marginal rate for federal income tax.

There would be another 0 to 8% for municipal tax (this depends on the municipality, I assume similar to state taxes?).

Then there’s 20% (employer contribution), and another 13% (employee contribution) for “rsz”, state mandated health insurance.

Sales tax is 21% on most goods, 6% on food.

Cries in european tears

I’m not sure the whole arab or asian world would agree. They’re still colonizing africa.

Why don’t you just use excel for that

I’ve this image in my mind of Del boy flogging cheese from a giant wheel in the back of his van

Maybe now they can shift to more ethical business models?

You can’t honestly expect that?

I’ve been using K-9 instead. What am I missing out on?

NPCs don’t mind

Stochastic parrot is a term describing your experience with AI.

I’ve an aunt like that, ready to engage in every conversation despite not understanding the topic. Regugitating whatever she saw on her tv, in her own words.

So in a sense, the current large language models do mimic some human’s behaviour.

Not necessarily, sometimes dimensionality reduction (the more common terminology used, for what is basically compression) is the explicit goal.

Can be used for outlier detection, similarity search, etc.

During training, you find a projection of the input, for example an image, to a smaller space, and then back to the original image. This is referred to as encoding and decoding. The error fuction would be a measure of how similar the in- and output images are.

AI does work like that.

With (variational) auto-encoders, it’s very explicit.

With shallow convolutional neural networks, it’s fun to visualize the trained kernel weights, as they often return an abstract, to me dreamlike, representations of the thing being trained for. Although derived through a different method, search for “eigenfaces” as an example of what I mean.

In the recent hype model architecture, attention and transformers, the encoded state can be thought of as a compressed version of it’s input. But human interpretation of those values is challenging.

I’m self employed.

Found out I’m depressed, decided to reduce work to 3 days a week.

Raised prices to reduce clients. Turns out I now make more, doing less work. It is what it is.