Transformer Explainer
Project record
Maintained by Polo Club of Data Science, Georgia Tech126
Transformer Explainer is an open-source interactive visualization that runs the GPT-2 (small) model, which has 124 million parameters, in the web browser. Users enter their own text and follow how the Transformer's components and operations combine to predict the next token. It was created by the Polo Club of Data Science at Georgia Tech and is described in a CHI 2026 paper.134
- Website: Live tool (external site: poloclub.github.io)
- Repository: Repository (poloclub/transformer-explainer) (external site: github.com)
- Paper: Paper: Transformer Explainer (arXiv 2408.04619) (external site: arxiv.org)
- License: LICENSE (MIT) (external site: github.com)
Availability and license
Overall availability
The tool runs in the browser at poloclub.github.io/transformer-explainer, and the source code is public on GitHub.13
Availability is separate from permission: read the license before using or redistributing.
Component reuse rights
- weights
- Unknown — no complete fact-level rights review
- code
- Reviewed qualifying license recorded — check scope and conditions
- data
- Unknown — no complete fact-level rights review
- documentation
- Unknown — no complete fact-level rights review
No complete system-rights review is recorded for this release.
Public materials checklist
| Item | Status | Notes and evidence |
|---|---|---|
| Source codeIs the source code publicly readable? | Public | Public GitHub repository under the MIT License.12 |
| DocumentationIs user documentation published? | Public | The README gives an overview and local setup steps, the live site includes written explanations of each Transformer component, and the design is described in the CHI 2026 paper.134 |
| InstallationAre installation instructions or packages publicly available? | Public | No installation is needed to use the hosted tool. To run it locally, the README lists Node.js v20 or later and npm v10 or later, then "npm install" and "npm run dev".1 |
| Supported platformsAre supported operating systems or hardware documented? | Partial | The tool runs in a web browser; the README does not list supported browsers or devices.1 |
| Release statusAre versioned releases published? | Partial | The GitHub releases page lists a single release, v0.0.1, from June 11, 2024, and no later numbered releases.5 |
What it is useful for
Learning how Transformer-based text-generation models work without installing anything: the paper describes an overview of the model's data flow with explanations that reveal mathematical detail on demand, and lets learners change the input and sampling settings such as temperature and see next-token predictions update.43
Run and use notes
- The README's local setup clones the repository, runs "npm install" and "npm run dev", and serves the tool at http://localhost:5173.1
Organization context
U.S. eligibility
Eligible · basis: U.S.-governed project
The repository sits in the GitHub organization of the Polo Club of Data Science at Georgia Tech, its MIT License names the Polo Club of Data Science as copyright holder, and its README says the tool was created at the Georgia Institute of Technology. Georgia Tech is a public research university in Atlanta, Georgia.6217
Sources
This listing is not an endorsement, a safety assessment, or a federal approval.