# poloclub/transformer-explainer

> Interactive visualization of how Transformer models work: runs a live GPT-2 in the browser and shows how internal components predict the next tokens.

- Magnitude: 5.0 out of 10 — Breakout
- Stars: 8,719 total · +9 stars today, ≈ 28 by evening
- Star trust: star growth looks organic
- Category: Learning & lists · Language: JavaScript · License: MIT · Created: 2024-05-16 · Last push: 2026-06-06
- GitHub: https://github.com/poloclub/transformer-explainer · Homepage: https://poloclub.github.io/transformer-explainer/ · Page: https://gitnova.dev/en/r/poloclub/transformer-explainer

## Useful for

- Learn how a Transformer works using a live GPT-2 in the browser
- Test with your own text how the model predicts the next tokens
- Walk through attention and other internal components step by step

## Why it’s here

- 9 stars so far today, about 28 expected by the end of the day. The usual pace is 7 per day, so that's 4.2× as much.
- Over the last two days the pace is 10× that of the previous week and a half.
- The spike has held for 3 days in a row — not a one-off blip.
- GitHub Trending JavaScript today: #5, +54 stars.
- Hacker News: “Transformers Explained Visually” — 616 points, 1 day ago.
- Recent forks include notable developers: @millette (309 followers).

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 983
- Issues and pull requests: 87
- Watchers: 73
- Average over the last week: 22 per day
- Usual pace: 7 per day
- Stars in the last hour (measured): -1
- Latest release: v0.0.1 (2024-06-11)

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-08-25 … 2026-09-23: 8, 6, 3, 6, 2, 11, 4, 4, 6, 3, 8, 9, 6, 17, 11, 9, 6, 3, 2, 3, 10, 7, 3, 8, 3, 2, 6, 39, 69, 9

## Hacker News

- Transformers Explained Visually — 616 points, 88 comments: https://news.ycombinator.com/item?id=49792342

## Spotted in now

- GitHub Trending JavaScript today: #5, +54 stars

## Similar by description

1. **FareedKhan-dev/train-llm-from-scratch** — 4.0 · Peaking · Language models · Python · +19 stars today, ≈ 48 by evening
   Educational project: a from-scratch PyTorch Transformer plus a full LLM training pipeline — from raw text through SFT, reward modeling, PPO/DPO/GRPO to chat, without transformers, trl or peft.
   Full card: https://gitnova.dev/en/r/FareedKhan-dev/train-llm-from-scratch.md
2. **petergpt/transformer-architecture** — 0.8 · Steady · Learning & lists · JavaScript · +0 stars today, ≈ 0 by evening
   Interactive 3D visualization of the Transformer and DeepSeek V4.1 Flash architectures, showing token flow, attention heads, expert routing and caching. An educational JavaScript project with no build step or API keys.
   Full card: https://gitnova.dev/en/r/petergpt/transformer-architecture.md
3. **Binaire-0101/FRZi-inference** — 0.0 · Quiet · Language models · +0 stars today, ≈ 0 by evening
   FRZi-inference is an inference engine for running machine learning models. It is designed to execute pre-trained models and obtain predictions.
   Full card: https://gitnova.dev/en/r/Binaire-0101/FRZi-inference.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-09-23 12:53 UTC, updated every 30 minutes.