4 ms·
Great stuff! I wrote an interactive blogpost that walks through the code and visualizes it: https://growingswe.com/blog/microgpt https://growingswe.com/blog/mic
by growingswe 7mo ago
Great stuff! I wrote an interactive blogpost that walks through the code and visualizes it: https://growingswe.com/blog/microgpt https://growingswe.com/blog/microgpt
- spinningslate 7mo agoThat’s beautifully done, thanks for posting. As helpful again to an ML novice like me as Karpathy’s original.
- hei-lima 7mo agoGreat!
- evntdrvn 7mo agoreally nice, thanks
- evntdrvn 7mo agoYou should totally submit that to HN as an article, if you haven't already.
- dang 7mo agoWe've put https://news.ycombinator.com/item?id=47205208 https://news.ycombinator.com/item?id=47205208 in the second-chance pool (https://news.ycombinator.com/pool https://news.ycombinator.com/pool, explained at https://news.ycombinator.com/item?id=26998308 https://news.ycombinator.com/item?id=26998308), so it will get a random placement on HN's front page.
- joenot443 7mo agoThis is awesome! Normally I'm pretty critical of LLM-assisted-blogging, but this one's a real winner.
- O4epegb 7mo ago> By the end of training, the model produces names like "kamon", "karai", "anna", and "anton". None of them are copies from the dataset. All 4 are in the dataset, btw
- mym1990 7mo agoThis is likely because the blog is AI generated and keys off this point from Karpathy: "As a preview, by the end of the script our model will generate (“hallucinate”!) new, plausible-sounding names.", so the LLM just repackaged that into something that is obviously wrong, which is kind of ironic.