Writing

Chapter 1: Tokenization — Where Language Becomes Discrete
April 3, 2026 · AI/ML · Thinking in Transformers
Why tokenization is the first information bottleneck in LLMs, shaping compression, context length, multilingual behavior, and symbolic performance.
- Tokenization
- BPE
- WordPiece
- SentencePiece
- Transformers
Chapter 0: The Design Space of Language Models
March 16, 2026 · AI/ML · Thinking in Transformers
A working model for understanding LLMs as compressed, lossy, differentiable databases of language.
- Transformers
- LLMs
- Language Models
- Architecture
- Design Space