nGPT's hypersphere constraint makes dot-product signal accumulate constructively under 4-bit quantization while noise averages out, enabling native low-precision training.
Robust implicit regularization via weight normalization.Information and Inference: A Journal of the IMA, 13(3):iaae022
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Normalized Architectures are Natively 4-Bit
nGPT's hypersphere constraint makes dot-product signal accumulate constructively under 4-bit quantization while noise averages out, enabling native low-precision training.