On the sizes of large subgraphs of the binomial random graph

We consider the binomial random graph $G(n,p)$, where p is a constant, and answer the following two questions.First, given $e(k)=p\left(\begin{array}{c}k\\ 2\end{array}\right)+O(k)$, what is the maximum k such that a.a.s. the binomial random graph $G(n,p)$ has an induced subgraph with k vertices and $e(k)$ edges? We prove that this maximum is not concentrated in any finite set (in contrast to the case of a small $e(k)$). Moreover, for every constant $C>0$, with probability bounded away from 0, the size of the concentration set is bigger than $C\sqrt{n/\ln ?n}$, and, for every ${\omega }_n\to \infty$, a.a.s. it is smaller than ${\omega }_n\sqrt{n/\ln ?n}$.Second, given $k>\epsilon n$, what is the maximum μ such that a.a.s. the set of sizes of k-vertex subgraphs of $G(n,p)$ contains a full interval of length μ? The answer is $\mu =\mathrm{\Theta }\left(\sqrt{(n?k)n\ln ?\left(\begin{array}{c}n\\ k\end{array}\right)}\right)$.