I think I have seen this in either McCoy, Ntantu: topological properties of spaces of continuous functions, and/or in Arhangel'skij's book on $C_p(X)$, the latter calls it $t_\mathbb{R}$ as a cardinal function IIRC. Something like real-tightness. So $t_{\mathbb{R}}(X) \le \tau$ iff ($f: X \rightarrow \mathbb{R}$ is continuous iff its restriction to all subsets of cardinality $\le \tau$ of $X$ is continuous).
I think it corresponds to some cardinal invariant of $C_p(X)$ and that this was the reason for its introduction. I don't have access to these books now, so I cannot check exactly, but I hope this helps anyway.