Show HN: Every number in a tiny GPT, from one forward pass to one step of RLHF

Developer manogyasingh released transformer-visualisation, a browser-based walkthrough of a toy GPT-2-like model that computes every matrix live and shows the exact math for each cell on hover, covering tokenization through sampling, one pretraining step, and one PPO iteration of RLHF. The project, built with Opus 5.5 and inspired in part by Brendan Bycroft's LLM visualization, was posted to Hacker News as a Show HN with 1 point and 0 comments.

Source: https://github.com/manogyasingh/transformer-visualisation https://github.com/manogyasingh/transformer-visualisation Made this with Opus 5.5 to understand the exact working of an LLM. It's a step by step walkthrough of a toy GPT-2 like model. Every matrix fits on screen, every number is computed live in the browser, and hovering over any cell shows the exact math of that step. It covers inference -- following the generation of one token from tokensiation to sampling, one training step of pretraining, and one PPO iteration of RLHF for post-training. Inspired in part by Brendan Bycroft's LLM visualization https://bbycroft.net/llm https://bbycroft.net/llm Comments URL: https://news.ycombinator.com/item?id=49991390 https://news.ycombinator.com/item?id=49991390 Points: 1 Comments: 0