Jupyter Notebook · MIT
raiyanyahya/how-to-train-your-gpt
Build a modern LLM from scratch. Every line commented. Explained like we are five.
★ 4.2k
Stars
504 forks · 1 open issues
+2.1/day
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
34.2
Momentum / 100
see the breakdown below
6mo ago
Created
last push 12d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 9.5 | +2.1/day, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 0.0 | last week ran 0.11× the monthly pace |
| Freshness | 15 | 14.2 | last push 12d ago; decays to zero at 180 days |
| Scale | 15 | 10.5 | 4.2k stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 34.2 | sum of the above |
Detail
| Topics | attention-mechanismdeep-learningeducationalfrom-scratchgptlanguage-modelllamallmmachine-learningnatural-language-processingpythonpytorchtokenisationtransformerstutorial |
| Matched on | Learning & Roadmaps (tutorial) |
| Short window | +15 stars over 7 days = 2.1/day |
| Long window | +512 stars over 26 days = 19.7/day |
| Lifetime average | 22.45 stars/day over 186 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |