Pinned Loading

1. attention-is-all-you-need attention-is-all-you-need Public

   From-scratch PyTorch implementation of multi-head attention with a GPT-2 head-behavior analysis measuring the entropy gradient across layers that the original paper never quantified.

   Python

2. llama-2 llama-2 Public

   Code-first implementation and visualizations of Llama 2 internals RoPE and GQA from scratch in Python.

   Python

3. mixtral-of-experts mixtral-of-experts Public

   From-scratch implementation and experiments for Mixtral of Experts (MoE) includes router analysis, isoFLOP scaling, and expert usage artifacts.

   Python

4. ARLibNavigation ARLibNavigation Public

   An AR-powered mobile application designed to help students navigate university libraries efficiently and locate books with ease.

   C++

5. FIFA-Player-Value-Prediction FIFA-Player-Value-Prediction Public

   Machine learning project using the FIFA dataset to predict football players' market value based on various attributes such as performance, age, and potential. Implemented in R with data preprocessi…

   R

6. mslogistics-fyp mslogistics-fyp Public

   C#