ajheshbasnet/LLaMA-2-from-Scratch

llama-2 from scratch — a clean, educational pytorch implementation of the llama-2 transformer architecture. features grouped query attention (gqa), rotary position embeddings (rope), kv caching, and swiglu activation. run easily on google colab and experiment with your own datasets.

10
/ 100
Experimental

No commits in the last 6 months.

No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 1 / 25
Maturity 7 / 25
Community 0 / 25

How are scores calculated?

Stars

1

Forks

Language

Python

License

Last pushed

Oct 01, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/ajheshbasnet/LLaMA-2-from-Scratch"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.