ggml-org/llama.cpp b12008

on GitHub

gemma3n, gemma4 : create token_embd before the tied output (#31394)

The duplicated output looks up token_embd in its buffer context, so
token_embd must exist first. With the old order a load where input and
output share a buffer type (CPU-only, -ngl 0) held the table twice:
gemma-4-E2B-it Q4_K_M, no mmap, CPU model buffer 3163.73 -> 2947.73 MiB.