import sys
import pypyjit


pypyjit.set_param(
    "threshold=1,"
    "function_threshold=1,"
    "trace_eagerness=1,"
    "decay=0"
)


W = 63

# r bit 左シフトするときに、
# 63bit の範囲に残る下位部分を取り出すための mask
LOW_MASK = [0] * W

r = 1
while r < W:
    LOW_MASK[r] = (1 << (W - r)) - 1
    r += 1


def solve() -> None:
    data = list(map(int, sys.stdin.buffer.read().split()))

    pos = 0
    T = data[pos]
    pos += 1

    out = [""] * T

    tc = 0
    while tc < T:
        N = data[pos]
        S = data[pos + 1]
        pos += 2

        # +1 は最上位 word からの carry 用
        top_word = S // W
        bits = [0] * (top_word + 2)

        # 部分和 0
        bits[0] = 1

        end = pos + N

        while pos < end:
            a = data[pos]
            pos += 1

            if a > S:
                continue

            q = a // W
            r = a - q * W

            # x + a <= S となり得る最大 source word
            src = (S - a) // W

            if r == 0:
                while src >= 0:
                    bits[src + q] |= bits[src]
                    src -= 1

            else:
                low_mask = LOW_MASK[r]
                rr = W - r

                while src >= 0:
                    x = bits[src]

                    dst = src + q

                    # dst word の bit r..62
                    bits[dst] |= (x & low_mask) << r

                    # 次の word の bit 0..r-1
                    bits[dst + 1] |= x >> rr

                    src -= 1

        # S 以下で最も大きい立っている bit を探す
        wi = top_word
        rb = S - wi * W

        x = bits[wi] & ((1 << (rb + 1)) - 1)

        if x:
            ans = wi * W + x.bit_length() - 1
        else:
            wi -= 1

            while bits[wi] == 0:
                wi -= 1

            ans = wi * W + bits[wi].bit_length() - 1

        out[tc] = str(ans)
        tc += 1

    sys.stdout.write("\n".join(out))


if __name__ == "__main__":
    solve()