Distillation and model compression · MLD-0808
NF4
A four-bit quantization data type designed around quantiles of a normal distribution.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Store base-model weights in QLoRA-style training.
Original caveat
Its advantage depends on weight distribution and implementation.
Context and aliases
Level: Intermediate. Stages: Training · Adaptation · Deployment.
Tags: DL · SLM · LLM.
Aliases: 4-bit NormalFloat.
Related concepts
Concepts linking here
Original references
- Distilling the Knowledge in a Neural Network ↗ · Consulted
- PEFT: LoRA conceptual guide ↗ · Consulted
- PEFT: quantization guide ↗ · Consulted
- Original reference URL ↗