This appendix provides standard definitions for the core terminology used throughout the book *Born*.
Tensors and Data Structures
| Term | Definition |
|------|------------|
| Tensor | A general term for multi-dimensional arrays; represented in Born as Tensor[T, B] |
| Shape | The list of lengths of a tensor along each axis |
| Stride | The memory offset required to move one element along each axis |
| Broadcasting | Rules for automatically expanding smaller shapes to match larger ones |
Compute Backends
| Term | Definition | |------|------------| | Backend | A concrete implementation that executes tensor operations (CPU/WebGPU) | | Lazy evaluation | Deferring operation execution and submitting them in batches to hide communication latency | | WGSL | WebGPU Shading Language | | Workgroup | The basic unit of compute dispatch in WebGPU |
Automatic Differentiation
| Term | Definition | |------|------------| | Automatic differentiation | Algorithms that compute function gradients automatically | | Gradient tape | A data structure that records the sequence of forward operations | | Backpropagation | Propagating gradients backward from the output loss through the computation graph |
Neural Networks
| Term | Definition | |------|------------| | Module | The basic building block of neural networks | | Multi-head attention (MHA) | Multiple attention heads computed in parallel | | Grouped-query attention (GQA) | Multiple query heads sharing key-value heads | | Residual connection | A shortcut connection that adds the input directly to the output |
Models and Deployment
| Term | Definition | |------|------------| | Checkpoint | A snapshot of model state saved during training | | GGUF | Georgi Gerganov Universal Format, a model format for LLMs | | KV-Cache | Caching key-value tensors to avoid repeated computation | | Quantization | Converting parameters from high precision to low precision |
---
📘 This is Appendix C (part 4) of *Born*, a technical book being published in serial form.