# poloclub/transformer-explainer

Transformer Explained Visually: Learn How LLM Transformer Models Work with Interactive Visualization

Repository: https://github.com/poloclub/transformer-explainer
Canonical: https://ross.abutalabs.com/products/transformer-explainer
Homepage: https://poloclub.github.io/transformer-explainer/
Language: JavaScript
License: MIT
License Family: permissive
Topics: generative-ai, gpt, deep-learning, langauge-model, llm, visualization
Last push: 2026-06-06T12:08:04+00:00

## Health v2 (maintenance only)
Score: 53/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 86, release rhythm 8, longevity 59
- inputs: {"age_days": 839, "days_push": 88, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8462, forks 942 (observed 2026-08-28T04:10:21.298512+00:00)

## What it is
An interactive web-based visualization tool that runs a live GPT-2 model in the browser to teach how Transformer-based LLMs work. Users can experiment with their own text and observe internal Transformer components like attention and next-token prediction in real time.

## Use cases
- learn how transformer models work interactively
- visualize self-attention in GPT-2
- understand next-token prediction
- teach LLM architecture to students
- experiment with temperature and top-k sampling
- run a small language model in the browser

## When to choose
- you want an intuitive, interactive introduction to Transformer and GPT internals
- you are teaching or self-studying how LLMs generate text
- you want to run a live model without installing anything, right in the browser

## When to avoid
- you need to train or fine-tune Transformer models
- you need state-of-the-art LLM inference performance
- you need a production NLP library rather than an educational tool

## Facets
- artifact type: learning-resource
- maturity: active
- function: data-visualization, machine-learning
- domain: deep-learning, large-language-models, education, data-visualization
- platform: browser
- tags: interactive-visualization, transformer, gpt-2, llm-education, in-browser-inference, svelte, education, nodejs, web-server

## Member repositories
- poloclub/transformer-explainer (main) score 53

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:21.298512+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:27:12.304164+00:00, confidence not recorded.
  - readme: https://github.com/poloclub/transformer-explainer (fetched 2026-08-28T04:10:21.298512+00:00, sha 5fe43cefde66)
  - homepage: https://poloclub.github.io/transformer-explainer/ (fetched 2026-08-29T08:27:15.436107+00:00, sha 81afc611fa9f)
- Data as of 2026-08-30T08:39:29.467469+00:00.