I recommend asking a friend who's a better writer and mathematician than Claude Code to help you reorganize the paper so that there are no gaps in the argumenta

I recommend asking a friend who's a better writer and mathematician than Claude Code to help you reorganize the paper so that there are no gaps in the argumentation and incorrect statements like "For a purely causal transformer without residuals, the gradient routed from the final token L to an earlier token j after H layers is given by the bottom row of the exponential Cesàro Matrix M^H" are replaced with mathematically correct descriptions. Also have them check your experiments, because the description doesn't inspire confidence your (Claude's) implementation isn't flawed in ways that invalidate your results. In particular, "our experimental code utilizes a highly efficient one-pass scalar-probe surrogate" sounds fishy.