in a random graph setting with $n$ vertices and edge-probability $p$ I want to show that $\binom n4p^6\to 0$ as $pn^{2/3}\to 0$. To check this is even true, I submitted this Wolfram|Alpha query.
Now why does Wolfram|Alpha calculate the limit for $n\to pn^{2/3}$ rather than $pn^{2/3}\to 0$? Is it just a misinterpretation, or is there a mathematical connection I fail to see? Surprisingly it yields the desired result ...
I also don't quite understand what Wolfram|Alpha does in the next line, and I am not even sure how to calculate the limit manually, so I'd be grateful for any insight!
Use $pn^{2/3}=\varepsilon$ and then $\varepsilon\to0$.
Wolfram|Alpha