diff --git a/main.tex b/main.tex
index 8c3ce83..9ff0d82 100644
--- a/main.tex
+++ b/main.tex
@@ -302,7 +302,7 @@ Sequence length & 2,048 tokens \\
 \centering
 \includegraphics[width=0.50\linewidth]{data/loss_compare.png}
 \caption{Training loss curves: GPT-2 residual scaling vs.~Kaiming
-initialization over $\sim$20B tokens.}
+initialization over $\sim$5B tokens.}
 \label{fig:loss}
 \end{figure}