NHacker Next
  • new
  • past
  • show
  • ask
  • show
  • jobs
  • submit
Microgpt explained interactively (growingswe.com)
politelemon 2 hours ago [-]
> 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/...

ayhanfuat 1 hours ago [-]
You are absolutely right. The whole post reads like AI generated.
jsheard 54 minutes ago [-]
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.

re 1 hours ago [-]
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.

butterisgood 1 hours ago [-]
ISWYDT
1 hours ago [-]
growingswe 53 minutes ago [-]
Thanks, will fix
malnourish 29 minutes ago [-]
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.
windowshopping 1 hours ago [-]
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.

nimbus-hn-test 2 hours ago [-]
[dead]
Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact
Rendered at 22:13:25 GMT+0000 (Coordinated Universal Time) with Vercel.