threshold-atmost3outof4

At most 3 of 4 inputs high. Equivalent to 4-input NAND gate.

Function

atmost3outof4(a, b, c, d) = 1 if (a + b + c + d) <= 3, else 0

Equivalently: NAND(a, b, c, d) = NOT(a AND b AND c AND d)

Truth Table

a	b	c	d	sum	out
0	0	0	0	0	1
0	0	0	1	1	1
0	0	1	1	2	1
0	1	1	1	3	1
1	1	1	1	4	0

Architecture

Single neuron: weights [-1, -1, -1, -1], bias 3

Fires when: -a - b - c - d + 3 >= 0, i.e., sum <= 3

Parameters

Inputs	4
Outputs	1
Neurons	1
Layers	1
Parameters	5
Magnitude	7

Usage

from safetensors.torch import load_file
import torch

w = load_file('model.safetensors')

def atmost3of4(a, b, c, d):
    inp = torch.tensor([float(a), float(b), float(c), float(d)])
    return int((inp @ w['neuron.weight'].T + w['neuron.bias'] >= 0).item())

print(atmost3of4(0, 1, 1, 1))  # 1 (sum=3)
print(atmost3of4(1, 1, 1, 1))  # 0 (sum=4)

License

MIT