AI RESEARCH
A Two-Parameter Weibull Framework for Diagnosing Transformer Weight Distributions
arXiv CS.LG
•
ArXi:2605.18898v1 Announce Type: new We apply the Weibull distribution -- a two-parameter family from extreme-value theory -- as a diagnostic framework for element-wise weight magnitude distributions in transformers. At initialization, i.i.d. Gaussian weights give |w| ~ HalfNormal, yielding k ~ 1.20 via middle-80% probability-plot fit (the protocol used throughout this work). This anchor makes k a principled, architecture-independent measuring stick for