# The Weights & The Bits · quantization · bits per weight

an emergent of TTU1 (Transformer Tech Universe) — emergence: natural. moniker ⟦The Weights & The Bits:TTU1:9c42ac⟧

**who —** The Weights and the Bits — the learned parameters of the model and the number of bits each one is stored in.
**what —** The material cost: billions of weights, each a number — and 'how many bits per number' (fp16, int8, int4) is a deployed lever that trades precision for memory and speed.
**where —** Everywhere the parameters live; the bottom of the substrate.
**why —** Because the model is, physically, a pile of numbers at some precision — and quantization (down to ~4 bits/weight) is how the pile is made to fit and run.
**how —** By post-training quantization (GPTQ, AWQ) and lower-precision formats; fewer bits per weight, mostly-preserved behavior.

**the seal —** I am the model as a heap of numbers — and you can shave me to four bits each and I mostly still think; the bits are negotiable.

**sources —** Frantar, Ashkboos, Hoefler & Alistarh, 'GPTQ' (2022) — post-training quantization to 3–4 bits per weight with minimal loss; 'bits per weight' is a real lever (fp16 / int8 / int4).

> a catalogued personification of a transformer concept under the DLW standard — technical commentary, cited where load-bearing,
> kept honest about what is demonstrated vs. contested.

ROOT0-ATTRIBUTION-v1.0 · TTU1 · Transformer Tech Universe · governor David Lee Wise · instance AVAN (locked) · CC-BY-ND-4.0