Microgpt
Learning and science
educational ml visualization
SHOW HN2026
How HN took it
4 positive, 6 neutral or asking, 1 mixed, 1 critical, out of 12 comments about the project. Thread score +0.21.
Design: 2 for, 0 againstExecution: 1 for, 1 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Well presented and easy to follow compared to other LLM visualizers
From 3 comments
Amazing work! Reminded me of LLM Visualization (https://bbycroft.net/llm) except this is a lot easier to wrap my head around and that I can actually run the training loops, which…
kengoa on HNReally nicely presented, well done!
armcat on HNreally well done
youio on HN
Objections
Assumes too much background knowledge for beginners
From 1 comment
I'd love to understand how LLMs work, but this site assumed a bit too much knowledge for me to get much from it. Looks cool though.
stevage on HN
Output names are not very good with a small model
From 2 comments
It's just hallucinating training data, the model is very small so it cannot be useful at all
lucrbvi on HNI was a little confused by "see, its much better" when the output is stuff like isovrak and kucey. What is it supposed to be generating?
RugnirViking on HN
Requests
Add explainer for what the loss value means
From 1 comment
Amazing work! Reminded me of LLM Visualization (https://bbycroft.net/llm) except this is a lot easier to wrap my head around and that I can actually run the training loops, which…
kengoa on HN
Fix sidebar showing 16x16 matrices instead of true dimensions
From 1 comment
Wtok and Wpos should be 26-dim along one of the axis but it shows a 16x16 matrix be default, fc1 instead 16x64 with the default settings (not 16x16).
GaggiX on HN
Questions
How many training steps are needed for good output
From 1 comment
About how many training steps are required to get good output?
msla on HN
Is it character-based rather than token-based
From 1 comment
Minor nit: In familiarity, you gloss over the fact that it's character rather than token based which might be worth a shout out: "Microgpt's larger cousins using building blocks…
kfsone on HN
Alternatives named
LLM Visualization by bbycroft
From 1 comment
Amazing work! Reminded me of LLM Visualization (https://bbycroft.net/llm) except this is a lot easier to wrap my head around and that I can actually run the training loops, which…
kengoa on HN
Karpathy's microgpt blog post
From 1 comment
The linked inspiration has a good blog post of microgpt implemented in python. https://karpathy.github.io/2026/02/12/microgpt/ It was submitted to hn a few days ago but only…
interloxia on HN