#include <algorithm>
#include <atcoder/all>
#include <bitset>
#include <cassert>
#include <cmath>
#include <ctime>
#include <functional>
#include <iomanip>
#include <iostream>
#include <map>
#include <queue>
#include <random>
#include <set>
#include <stack>
#include <string>
#include <tuple>
#include <unordered_map>
#include <unordered_set>
#include <utility>
#include <vector>
using namespace std;
using namespace atcoder;
typedef long long ll;
#define rep(i, n) for (int i = 0; i < (int)(n); i++)
#define repr(i, n) for (int i = (int)(n) - 1; i >= 0; i--)
#define repk(i, k, n) for (int i = k; i < (int)(n); i++)
#define all(v) v.begin(), v.end()
#define mod1 1000000007
#define mod2 998244353
#define mod3 100000007
#define vi vector<int>
#define vs vector<string>
#define vc vector<char>
#define vl vector<ll>
#define vb vector<bool>
#define vvi vector<vector<int>>
#define vvc vector<vector<char>>
#define vvl vector<vector<ll>>
#define vvb vector<vector<bool>>
#define vvvi vector<vector<vector<int>>>
#define vvvl vector<vector<vector<ll>>>
#define pii pair<int, int>
#define pil pair<int, ll>
#define pli pair<ll, int>
#define pll pair<ll, ll>
#define vpii vector<pair<int, int>>
#define vpll vector<pair<ll, ll>>
#define vvpii vector<vector<pair<int, int>>>
#define vvpll vector<vector<pair<ll, ll>>>

template <typename T>
void debug(T e) {
    cerr << e << endl;
}

template <typename T>
void debug(vector<T> &v) {
    rep(i, v.size()) { cerr << v[i] << " "; }
    cerr << endl;
}

template <typename T>
void debug(vector<vector<T>> &v) {
    rep(i, v.size()) {
        rep(j, v[i].size()) { cerr << v[i][j] << " "; }
        cerr << endl;
    }
}

template <typename T>
void debug(vector<pair<T, T>> &v) {
    rep(i, v.size()) { cerr << v[i].first << " " << v[i].second << endl; }
}

template <typename T>
void debug(set<T> &st) {
    for (auto itr = st.begin(); itr != st.end(); itr++) {
        cerr << *itr << " ";
    }
    cerr << endl;
}

template <typename T>
void debug(multiset<T> &ms) {
    for (auto itr = ms.begin(); itr != ms.end(); itr++) {
        cerr << *itr << " ";
    }
    cerr << endl;
}

template <typename T>
void debug(map<T, T> &mp) {
    for (auto itr = mp.begin(); itr != mp.end(); itr++) {
        cerr << itr->first << " " << itr->second << endl;
    }
}

void debug_out() { cerr << endl; }

template <typename Head, typename... Tail>
void debug_out(Head H, Tail... T) {
    cerr << H << " ";
    debug_out(T...);
}

using mint = modint998244353;

void debug_mint1(vector<mint> &vec) {
    for (int i = 0; i < vec.size(); i++) {
        cerr << vec[i].val() << " ";
    }
    cerr << endl;
}

void debug_mint2(vector<vector<mint>> &vec) {
    for (int i = 0; i < vec.size(); i++) {
        for (int j = 0; j < vec[i].size(); j++) {
            cerr << vec[i][j].val() << " ";
        }
        cerr << endl;
    }
}

ll my_pow(ll x, ll n, ll mod) {
    // 　繰り返し二乗法．x^nをmodで割った余り．
    ll ret;
    if (n == 0) {
        ret = 1;
    } else if (n % 2 == 1) {
        ret = (x * my_pow((x * x) % mod, n / 2, mod)) % mod;
    } else {
        ret = my_pow((x * x) % mod, n / 2, mod);
    }
    return ret;
}

ll inv(ll x, ll mod) { return my_pow(x, mod - 2, mod); }

vector<ll> fwht(vector<ll> &vec, ll K){
    ll h = 1;
    vector<ll> vec_cpy(vec.size());
    for (ll i = 0; i < vec.size(); i++){
        vec_cpy[i] = vec[i];
    }
    for (ll k = 0; k < K; k++){
        for (ll i = 0; i < vec.size(); i += h * 2){
            for (ll j = i; j < i + h; j++){
                ll v1 = vec_cpy[j];
                ll v2 = vec_cpy[j + h];
                vec_cpy[j] = (v1 + v2) % mod2;
                vec_cpy[j + h] = (v1 + mod2 - v2) % mod2;
            }
        }
        h *= 2;
    }
    return vec_cpy;
}

vector<ll> xor_convolution(vector<ll> &a, vector<ll> &b, ll K){
    vector<ll> a_cpy = fwht(a, K);
    vector<ll> b_cpy = fwht(b, K);
    vector<ll> c(a.size());
    for (ll i = 0; i < a.size(); i++){
        c[i] = (a_cpy[i] * b_cpy[i]) % mod2;
    }
    vector<ll> c_cpy = fwht(c, K);
    for (ll i = 0; i < c.size(); i++){
        c_cpy[i] = (c_cpy[i] * inv(1 << K, mod2)) % mod2;
    }
    return c_cpy;
}

int main() {
    ll N;
    cin >> N;
    vector<ll> A(N);
    rep(i, N) cin >> A[i];

    // この部分の計算量削減
    mint a2 = 0;
    /*for (ll i = 0; i < N; i++){
        for (ll j = 0; j < N; j++){
            for (ll k = 0; k < N; k++){
                a2 -= mint(8) * min((A[i] ^ A[j]), (A[i] ^ A[k]));
            }
        }
    }*/

    mint ans = 0;

    ll L = (1 << 18);
    ll B = 18;
    vector<vector<vector<vector<ll>>>> vals(B, vector<vector<vector<ll>>>(2, vector<vector<ll>>(B, vector<ll>(2, 0))));
    for (ll i = 0; i < N; i++){
        for (ll j = 0; j < B; j++){
            for (ll k = 0; k < B; k++){
                vals[j][(A[i] >> j) & 1][k][(A[i] >> k) & 1]++;
            }
        }
    }

    vector<vector<vector<ll>>> trie(2 * L, vector<vector<ll>>(B, vector<ll>(2, 0)));
    for (ll i = 0; i < N; i++){
        for (ll j = 0; j < B; j++){
            trie[A[i] + L][j][(A[i] >> j) & 1]++;
        }
    }
    for (ll j = 0; j < B; j++){
        for (ll k = 0; k < 2; k++){
            for (ll i = L - 1; i >= 1; i--){
                trie[i][j][k] += trie[2 * i][j][k];
                trie[i][j][k] += trie[2 * i + 1][j][k];
            }
        }
    }

    // この情報から和を求められる
    for (ll i = 1; i < L; i++){
        ll msb = -1;
        for (ll j = 0; j < B; j++){
            if ((i >> j) & 1) msb = j;
        }
        msb = B - 1 - msb;
        for (ll j = 0; j < B; j++){
            mint pj = mint(1LL << j);
            mint s0 = mint(trie[i * 2][j][0] + trie[i * 2][j][1]);
            mint s1 = mint(trie[i * 2 + 1][j][0] + trie[i * 2 + 1][j][1]);

            a2 += mint(2) * s0 * mint(vals[msb][1][j][0]) * mint(trie[2 * i + 1][j][1]) * pj;
            a2 += mint(2) * s0 * mint(vals[msb][1][j][1]) * mint(trie[2 * i + 1][j][0]) * pj;
            a2 += mint(2) * s1 * mint(vals[msb][0][j][0]) * mint(trie[2 * i][j][1]) * pj;
            a2 += mint(2) * s1 * mint(vals[msb][0][j][1]) * mint(trie[2 * i][j][0]) * pj;
        }
    }

    vector<ll> coef(L, 0);
    rep(i, N) coef[A[i]]++;

    vector<ll> coef2(L, 0);
    rep(i, L){
        coef2[i] = (coef[i] * coef[i]) % mod2;
    }

    vector<vector<ll>> val2(B, vector<ll>(2, 0));
    for (ll i = 0; i < L; i++){
        for (ll j = 0; j < B; j++){
            val2[j][(i >> j) & 1] += coef2[i];
            val2[j][(i >> j) & 1] %= mod2;
        }
    }
    for (ll i = 0; i < N; i++){
        for (ll j = 0; j < B; j++){
            mint pj = mint(1LL << j);
            a2 += pj * mint(val2[j][1 - ((A[i] >> j) & 1)]);
        }
    }

    a2 *= mint(mod2 - 8);


    vector<ll> conv = xor_convolution(coef, coef, B);
    conv[0] -= N;
    
    mint now_sum = 0;
    mint now_cnt = 0;
    for (ll i = 0; i < L; i++){
        ans += mint(conv[i]) * (mint(i) * now_cnt - now_sum);
        now_sum += i * conv[i];
        now_cnt += conv[i];
    }
    for (ll i = 0; i < L; i++){
        ans += (4 * conv[i] * i);
    }

    a2 += N * 8 * now_sum;
    
    cout << (ans * mint(2) - a2).val() << endl;
}