#include <iostream>
#include <algorithm>
#include <vector>

typedef std::pair<int, int> pii;

#define fi first
#define se second

using namespace std;

int sub_mod(int p, int q, int mod) {
    return (p <= q ? q - p : mod - p + q);
}

int main() {
    int n;
    cin >> n;
    vector<pii> nums(n);
    for (auto &el : nums) cin >> el.fi >> el.se;

    int dp[n][5][3];
    for (int i = 0; i < n; ++i) for (int j = 0; j < 5; ++j) for (int k = 0; k < 3; ++k) dp[i][j][k] = 0;

    dp[0][nums[0].fi % 5][nums[0].se % 3] = nums[0].fi;
    dp[0][nums[0].se % 5][nums[0].fi % 3] = nums[0].se;

    for (int i = 1; i < n; ++i) {
        int p1 = nums[i].fi % 5, p2 = nums[i].se % 5;
        int q1 = nums[i].fi % 3, q2 = nums[i].se % 3;
        for (int j = 0; j < 5; ++j) {
            for (int k = 0; k < 3; ++k) {
                dp[i][j][k] = max(dp[i][j][k], dp[i - 1][sub_mod(p1, j, 5)][sub_mod(q2, k, 3)] + nums[i].fi);
                dp[i][j][k] = max(dp[i][j][k], dp[i - 1][sub_mod(p2, j, 5)][sub_mod(q1, k, 3)] + nums[i].se);
            }
        }
    }

    int ans = 0;
    for (int j = 1; j <= 4; ++j) {
        for (int k = 1; k <= 2; ++k) {
            ans = max(ans, dp[n - 1][j][k]);
        }
    }
    cout << ans << '\n';
    return 0;
}