notes/

implementations/
  • jpeg compression/
  • karpathy 01 micrograd/
  • karpathy 02 bigram/
  • karpathy 03 mlp/
  • karpathy 04 batchnorm/
        • implementations/
        • raw/
        • wiki/
        • AGENTS

              Implementations

              May 14, 20261 min read

              Hands-on reimplementations with annotated notebooks.

              Builds — 5 folders

              Backpropagation & Autograd
              9 notebooks
              Scalar autograd engine, manual backprop, automated backward pass, and a minimal neural net library.
              Bigram Language Model
              5 notebooks + 2 helpers
              Character-level counting model, sampling, loss, smoothing, and an equivalent neural net.
              MLP / Makemore
              5 notebooks
              Scaling the bigram model to an MLP with embeddings, train/val/test splits, and experiments.
              BatchNorm
              1 notebook
              Training dynamics, activation statistics, gradient flow, initialization, and normalization.
              JPEG Compression
              1 notebook
              Colour conversion, 8x8 DCT blocks, quantization, zig-zag scanning, RLE, and Huffman coding.

              • No backlinks found

              Created with Quartz v4.5.2 © 2026