#include <bits/stdc++.h>
 
//#include <unistd.h>
//#include <iostream>
 
using namespace std;
 
#define DEBUG(x) cerr<<#x<<": "<<x<<endl;
#define DEBUG_VEC(v) cerr<<#v<<":";for(int i=0;i<v.size();i++) cerr<<" "<<v[i]; cerr<<endl;
#define DEBUG_MAT(v) cerr<<#v<<endl;for(int i=0;i<v.size();i++){for(int j=0;j<v[i].size();j++) {cerr<<v[i][j]<<" ";}cerr<<endl;}
typedef long long ll;
#define int ll
 
#define vi vector<int>
#define vl vector<ll>
#define vii vector< vector<int> >
#define vll vector< vector<ll> >
#define vs vector<string>
#define pii pair<int,int>
#define pis pair<int,string>
#define psi pair<string,int>
#define pll pair<ll,ll>
template<class S, class T> pair<S, T> operator+(const pair<S, T> &s, const pair<S, T> &t) { return pair<S, T>(s.first + t.first, s.second + t.second); }
template<class S, class T> pair<S, T> operator-(const pair<S, T> &s, const pair<S, T> &t) { return pair<S, T>(s.first - t.first, s.second - t.second); }
template<class S, class T> ostream& operator<<(ostream& os, pair<S, T> p) { os << "(" << p.first << ", " << p.second << ")"; return os; }
#define X first
#define Y second
#define rep(i,n) for(int i=0;i<(n);i++)
#define rep1(i,n) for(int i=1;i<=(n);i++)
#define rrep(i,n) for(int i=(n)-1;i>=0;i--)
#define rrep1(i,n) for(int i=(n);i>0;i--)
#define REP(i,a,b) for(int i=a;i<b;i++)
#define in(x, a, b) (a <= x && x < b)
#define all(c) c.begin(),c.end()
template<class T> bool chmax(T &a, const T &b) { if (a<b) { a = b; return 1; } return 0; }
template<class T> bool chmin(T &a, const T &b) { if (a>b) { a = b; return 1; } return 0; }
#define UNIQUE(v) v.erase(std::unique(v.begin(), v.end()), v.end());
const ll inf = 1000000001;
const ll INF = (ll)1e18 + 1;
const long double pi = 3.1415926535897932384626433832795028841971L;
#define Sp(p) cout<<setprecision(25)<< fixed<<p<<endl;
int dx[4] = {1, 0, -1, 0}, dy[4] = {0, 1, 0, -1};
int dx2[8] = { 1,1,0,-1,-1,-1,0,1 }, dy2[8] = { 0,1,1,1,0,-1,-1,-1 };
#define fio() cin.tie(0); ios::sync_with_stdio(false);
const ll MOD = 1000000007;
//const ll MOD = 998244353;
// #define mp make_pair
//#define endl '\n'

typedef pair<int, int> Pii;

#define FOR(i,n) for(int i = 0; i < (n); i++)
#define sz(c) ((int)(c).size())
#define ten(x) ((int)1e##x)

template<class T> T extgcd(T a, T b, T& x, T& y) { for (T u = y = 1, v = x = 0; a;) { T q = b / a; swap(x -= q * u, u); swap(y -= q * v, v); swap(b -= q * a, a); } return b; }
template<class T> T mod_inv(T a, T m) { T x, y; extgcd(a, m, x, y); return (m + x % m) % m; }
ll mod_pow(ll a, ll n, ll mod) { ll ret = 1; ll p = a % mod; while (n) { if (n & 1) ret = ret * p % mod; p = p * p % mod; n >>= 1; } return ret; }

template<int mod, int primitive_root>
class NTT {
public:
	int get_mod() const { return mod; }
	void _ntt(vector<ll>& a, int sign) {
		const int n = sz(a);
		assert((n ^ (n&-n)) == 0); //n = 2^k

		const int g = 3; //g is primitive root of mod
		int h = (int)mod_pow(g, (mod - 1) / n, mod); // h^n = 1
		if (sign == -1) h = (int)mod_inv(h, mod); //h = h^-1 % mod

		//bit reverse
		int i = 0;
		for (int j = 1; j < n - 1; ++j) {
			for (int k = n >> 1; k >(i ^= k); k >>= 1);
			if (j < i) swap(a[i], a[j]);
		}

		for (int m = 1; m < n; m *= 2) {
			const int m2 = 2 * m;
			const ll base = mod_pow(h, n / m2, mod);
			ll w = 1;
			FOR(x, m) {
				for (int s = x; s < n; s += m2) {
					ll u = a[s];
					ll d = a[s + m] * w % mod;
					a[s] = u + d;
					if (a[s] >= mod) a[s] -= mod;
					a[s + m] = u - d;
					if (a[s + m] < 0) a[s + m] += mod;
				}
				w = w * base % mod;
			}
		}

		for (auto& x : a) if (x < 0) x += mod;
	}
	void ntt(vector<ll>& input) {
		_ntt(input, 1);
	}
	void intt(vector<ll>& input) {
		_ntt(input, -1);
		const int n_inv = mod_inv(sz(input), mod);
		for (auto& x : input) x = x * n_inv % mod;
	}

	// 畳み込み演算を行う
	vector<ll> convolution(const vector<ll>& a, const vector<ll>& b){
		int ntt_size = 1;
		while (ntt_size < sz(a) + sz(b)) ntt_size *= 2;

		vector<ll> _a = a, _b = b;
		_a.resize(ntt_size); _b.resize(ntt_size);

		ntt(_a);
		ntt(_b);

		FOR(i, ntt_size){
			(_a[i] *= _b[i]) %= mod;
		}

		intt(_a);
		return _a;
	}
};

ll garner(vector<Pii> mr, int mod){
	mr.emplace_back(mod, 0);

	vector<ll> coffs(sz(mr), 1);
	vector<ll> constants(sz(mr), 0);
	FOR(i, sz(mr) - 1){
		// coffs[i] * v + constants[i] == mr[i].second (mod mr[i].first) を解く
		ll v = (mr[i].second - constants[i]) * mod_inv<ll>(coffs[i], mr[i].first) % mr[i].first;
		if (v < 0) v += mr[i].first;

		for (int j = i + 1; j < sz(mr); j++) {
			(constants[j] += coffs[j] * v) %= mr[j].first;
			(coffs[j] *= mr[i].first) %= mr[j].first;
		}
	}

	return constants[sz(mr) - 1];
}

typedef NTT<167772161, 3> NTT_1;
typedef NTT<469762049, 3> NTT_2;
typedef NTT<1224736769, 3> NTT_3;

//任意のmodで畳み込み演算 O(n log n)
vector<ll> int32mod_convolution(vector<ll> a, vector<ll> b,int mod){
	for (auto& x : a) x %= mod;
	for (auto& x : b) x %= mod;
	NTT_1 ntt1; NTT_2 ntt2; NTT_3 ntt3;
	auto x = ntt1.convolution(a, b);
	auto y = ntt2.convolution(a, b);
	auto z = ntt3.convolution(a, b);

	vector<ll> ret(sz(x));
	vector<Pii> mr(3);
	FOR(i, sz(x)){
		mr[0].first = ntt1.get_mod(), mr[0].second = (int)x[i];
		mr[1].first = ntt2.get_mod(), mr[1].second = (int)y[i];
		mr[2].first = ntt3.get_mod(), mr[2].second = (int)z[i];
		ret[i] = garner(mr, mod);
	}

	return ret;
}


signed main() {
	fio();
	int p;
	cin >> p;
	int n = 2000000;
	vl a(n);
	a[0] = 0;
	a[1] = 1;
	REP(i, 2, n) a[i] = (a[i - 1] * p % MOD + a[i - 2]) % MOD;

	//DEBUG_VEC(a);
	vl b = a;
	vl c = int32mod_convolution(a, b, MOD);

	int q;
	cin >> q;
	while (q--) {
		int s;
		cin >> s;
		s -= 2;
		cout << c[s] << '\n';
	}
}