diff --git a/transformer_sizing.ipynb b/transformer_sizing.ipynb
index 5fb60f8..53791ae 100644
--- a/transformer_sizing.ipynb
+++ b/transformer_sizing.ipynb
@@ -358,7 +358,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "This is not a bad estimate at all. I trained this model and it converged in roughly 4 days."
+    "This is not a bad estimate at all. I trained this model and it converged in roughly 4 days. Btw as a good reference for where 6ND comes from and some intuition around it I recommend [Dzmitry's post](https://medium.com/@dzmitrybahdanau/the-flops-calculus-of-language-model-training-3b19c1f025e4)."
    ]
   },
   {