alt.hn

3/1/2026 at 9:43:43 AM

Microgpt explained interactively

 https://growingswe.com/blog/microgpt

by growingswe

3/1/2026 at 10:31:44 PM

Is it becoming a thing to misspell and add grammatical mistakes on purpose to show that an LLM didn't write the blog post? I noticed several spelling mistakes in Karpathy's blog post that this article is based on and in this article.

by love2read

3/1/2026 at 8:15:48 PM

> By the end of training, the model produces names like "kamon", "karai", "anna", and "anton". None of them are copies from the dataset.

Hey, I am able to see kamon, karai, anna, and anton in the dataset, it'd be worth using some other names: https://raw.githubusercontent.com/karpathy/makemore/988aa59/...

by politelemon

3/1/2026 at 8:47:37 PM

You are absolutely right. The whole post reads like AI generated.

by ayhanfuat

3/1/2026 at 9:19:12 PM

The rate they are posting new articles on random subjects is also a pretty indicative of a content mill.

In 3 days they've covered machine learning, geometry, cryptography, file formats and directory services.

by jsheard

3/1/2026 at 9:11:23 PM

I didn't get that sense from the prose; it didn't have the usual LLM hallmarks to me, though I'm not enough of an expert in the space to pick up on inaccuracies/hallucinations.

The "TRAINING" visualization does seem synthetic though, the graph is a bit too "perfect" and it's odd that the generated names don't update for every step.

by re

3/1/2026 at 9:01:21 PM

ISWYDT

by butterisgood

3/1/2026 at 9:20:15 PM

Thanks, will fix

by growingswe

3/1/2026 at 10:16:18 PM

It says its tailored for beginners, but I don't know what kind of beginner can parse multiple paragraphs like this:

"How wrong was the prediction? We need a single number that captures "the model thought the correct answer was unlikely." If the model assigns probability 0.9 to the correct next token, the loss is low (0.1). If it assigns probability 0.01, the loss is high (4.6). The formula is − log ⁡ ( � ) −log(p) where � p is the probability the model assigned to the correct token. This is called cross-entropy loss."

by jmkd

3/1/2026 at 9:44:11 PM

I read through this entire article. There was some value in it, but I found it to be very "draw the rest of the owl". It read like introductions to conceptual elements or even proper segues had been edited out. That said, I appreciated the interactive components.

by malnourish

3/1/2026 at 10:19:46 PM

It started off nicely but before long you get

"The MLP (multilayer perceptron) is a two-layer feed-forward network: project up to 64 dimensions, apply ReLU (zero out negatives), project back to 16"

Which starts to feel pretty owly indeed.

I think the whole thing could be expanded to cover some more of it in greater depth.

by davidw

3/1/2026 at 9:10:41 PM

The part that eludes me is how you get from this to the capability to debug arbitrary coding problems. How does statistical inference become reasoning?

For a long time, it seemed the answer was it doesn't. But now, using Claude code daily, it seems it does.

by windowshopping

3/1/2026 at 10:29:48 PM

IMO your question is the largest unknown in the ML research field (neural net interpretability is a related area), but the most basic explanation is "if we can always accurately guess the next 'correct' word, then we will always answer questions correctly".

An enormous amount of research+eng work (most of the work of frontier labs) is being poured into making that 'correct' modifier happen, rather than just predicting the next token from 'the internet' (naive original training corpus). This work takes the form of improved training data (e.g. expert annotations), human-feedback finetuning (e.g. RLHF), and most recently reinforcement learning (e.g. RLVR, meaning RL with verifiable rewards), where the model is trained to find the correct answer to a problem without 'token-level guidance'. RL for LLMs is a very hot research area and very tricky to solve correctly.

by ferris-booler

3/1/2026 at 10:26:43 PM

Because it's not statistical inference on words or characters but rather stacked layers of statistical inference on ~arbitrarily complex semantic concepts which is then performed recursively.

by fc417fc802

3/1/2026 at 8:29:08 PM

[dead]

by nimbus-hn-test