#include <bits/stdc++.h>
using namespace std;

#define rep(i, a, b) for(int i = a; i < (b); ++i)
#define all(x) begin(x), end(x)
#define sz(x) (int)(x).size()
typedef long long ll;
typedef pair<int, int> pii;
typedef vector<int> vi;
template <typename T> using min_heap = priority_queue<T, vector<T>, greater<T>>;

void solve() {
	int n, k; cin >> n >> k;
	if (k > n/2) {
		cout << "NO\n";
		return;
	}

	cout << "YES\n";
	vector<int> p;
	for (int i = 1; i <= n - n%2; i++) {
		p.push_back(i-1);
	}

	int ans[n][n];
	memset(ans, 0, sizeof(ans));
	for (int i = 0; i < n/2; i++) {
		int col = min(i+1, k);
		for (int j = 0; j+1 < n/2; j++) {
			ans[p[j]][p[j+1]] = ans[p[j+1]][p[j]] = col;
			ans[p[j+n/2]][p[j+n/2+1]] = ans[p[j+n/2+1]][p[j+n/2]] = col;

			//cout << "EDGE " << p[j] << ' ' << p[j+1] << ' ' << col << '\n';
			//cout << "EDGE " << p[j+n/2] << ' ' << p[j+n/2+1] << ' ' << col << '\n';
		}
		ans[p[0]][p[n/2]] = ans[p[n/2]][p[0]] = col;
		//cout << "EDGE " << p[0] << ' ' << p[n/2] << ' ' << col << '\n';

		if (n % 2) {
			ans[p[0]][n-1] = ans[n-1][p[0]] = col;
			ans[p[n/2]][n-1] = ans[n-1][p[n/2]] = col;

			//cout << "EDGE " << p[0] << ' ' << n-1 << ' ' << col << '\n';
			//cout << "EDGE " << p[n/2] << ' ' << n-1 << ' ' << col << '\n';
		}
		rotate(p.begin(), p.begin()+1, p.begin()+(n - n%2));
	}

	for (int i = 1; i < n; i++) {
		for (int j = 0; j < i; j++) {
			cout << ans[i][j] << ' ';
		}
		cout << '\n';
	}
}

signed main() {
	ios::sync_with_stdio(0); cin.tie(0);
	
	int t; cin >> t;
	while (t--) {
		solve();
	}
	
	return 0;
}