Why is BitNet called 1.58-bit?
Three possible weight values require log2(3), approximately 1.585 bits, in an ideal encoding. Real model files also contain packing, scaling and other overhead.
Answered in
BitNet b1.58: What Ternary Weights Change—and What They Do NotBitNet trains with ternary weights to cut storage and arithmetic costs. Its results do not prove a universal energy multiplier or a 70B CPU-cache model.
Read the full analysisOther questions this article answers
More how ai actually works questions
- What are the three scenarios for the AI economy?
- What happens in the 25% Bull Case?
- What would trigger the 15% Bear Case crash?
- What does the 60% base case look like?
- Who are the winners in the 60% base case?
- Can any existing model be converted to BitNet?
- Does GraphRAG eliminate hallucinations?
- When is GraphRAG useful?
Every answer on Crashtech is written by the editor of the article it comes from — never auto-summarised. Browse all answers or the How AI Actually Works beat.