10units
38lectures
38live
P · Prerequisite — the vectors chapter
CPU. No GPU. Block the rest until this is done.- 01Tensors are the vectorsDrag two arrows. The rest of the course is this, with a shape.ready
- 02Matmul is the forceC = AB. Hover a cell. ~90% of GPU arithmetic.ready
- 03Softmax and cross-entropyLogits → distribution. Loss is surprise at the true id.ready
- 04Autograd — the chain ruleGreen forward. Amber back. One hop per node.ready
- 05Gradient descentClick the map. Batch is smooth. SGD jitters. η is the game.ready
- 06Bytes become idsBPE is compression. English vs Go vs YAML.ready
- 07A tiny net that actually fitsWatch the boundary curl. Turn the hidden layer off — it fails.ready
- 08Chat is a format, not an architectureRoles are tokens. Break the tags and the voice leaks.ready
D · Deep nets (refresh)
Refresh. Same sliders, shorter copy.- 01The neuronDashed line, solid bend. Switch tanh / ReLU / sigmoid.ready
- 02BackpropPress backward. Watch ∂L hop left.ready
- 03SGD, momentum, AdamSame saddle. Three paths. β→0 makes momentum = SGD.ready
- 04ConvolutionsHover a pixel. One kernel, every place.ready
- 05Sequences and vanishing gradientsg₀/g_T → 0 unless a gate holds ≈ 1.ready