News for hackers. Stuff that matters.
Re:Microgpt by teleforce on Sunday March 01, 2026 @01:21AM (#47204188)
Someone has modified microgpt to build a tiny GPT that generates Korean first names, and created a web page that visualizes the entire process [1].

Users can interactively explore the microgpt pipeline end to end, from tokenization until inference.

[1] English GPT lab:

https://ko-microgpt.vercel.app/

Re:Microgpt by camkego on Sunday March 01, 2026 @03:45PM (#47210484)
I have no affiliation with the website, but the website is pretty neat if you are learning LLM internals. It explains: Tokenization, Embedding, Attention, Loss & Gradient, Training, Inference and comparison to "Real GPT"

Pretty nifty. Even if you are not interested in the Korean language

Re:Microgpt by love2read on Sunday March 01, 2026 @06:03PM (#47211703)
By "modified" this person of course means that they swapped out the list of X0,000 names from English to Korean names. That is seemingly the only change.

The attached website is a fully ai-generated "visualization" based on the original blog post with little added.

Re:Microgpt by big-chungus4 on Monday March 02, 2026 @01:14AM (#47214458)
It's a good website and probably AI generated with some insane expensive model that us mere mortals are too poor to afford, thus it has a value
Re:Microgpt by sprobertson on Sunday March 01, 2026 @07:05PM (#47212218)
This kind of thing is pretty easy to do with a much leaner model https://docs.pytorch.org/tutorials/intermediate/char_rnn_gen...
Re:Microgpt by arijun on Monday March 02, 2026 @10:22AM (#47219112)
I assume the goal isn't to generate Korean names but to learn GPTs.
Re:Microgpt by sprobertson on Tuesday March 03, 2026 @05:49PM (#47240209)
Sure, but use the tool for the job IMO. GPTs are much more complex so should demonstrate a much more complex task.
Re:Microgpt by sahildeepreel on Monday March 02, 2026 @09:15AM (#47218228)
so impressive!