Let $ X_i $ denote the value of the die on the $ i^{th} $ roll. We want to find $P\big(\sum_{i=1}^{104} X_i\big) \leq 400 $.
As @lulu pointed out in the comments section, you want to use the CLT, i.e.
$$ \frac{\sum_{i=1}^{104}X_i - E(\sum_{i=1}^{104}X_i)}{\sqrt{Var(\sum_{i=1}^{104}X_i)}} \rightarrow Z $$
Since $E(X) = \frac{1}{6}\big(1 + 2 + 3 + 4 + 5 + 6\big) = \frac{7}{2}$ and $Var(X)=E(X^2)-(E(X))^2 = \frac{1}{6}\big(1^2 + 2^2 + 3^2 + 4^2 + 5^2 + 6^2\big) - \Big(\frac{7}{2}\Big)^2 = \frac{35}{12} $, we get:
$$P\Big(\sum_{i=1}^{104} X_i \leq 400\Big) = P\Bigg(\frac{\sum_{i=1}^{104}X_i - E(\sum_{i=1}^{104}X_i)}{\sqrt{Var(\sum_{i=1}^{104}X_i)}} \leq \frac{400 - E(\sum_{i=1}^{104}X_i)}{\sqrt{Var(\sum_{i=1}^{104}X_i)}}\Bigg) \approx P\Bigg(Z \leq \frac{400 - 104E(X)}{\sqrt{104Var(X)}}\Bigg) \approx P\Bigg(Z \leq \frac{400 - 104E(X)}{\sqrt{104Var(X)}}\Bigg) \approx P\Bigg(Z \leq \frac{400 - 104(7/2)}{\sqrt{104(35/12)}}\Bigg) \approx P(Z \leq 2.07) \approx 0.05 $$
Note: To improve the precision of your answer, you could also apply the continuity correction, which would mean calculating $P(\sum_{i=1}^{104} \leq 400.5)$ instead of $P(\sum_{i=1}^{104} \leq 400)$.