README.md

---
license: mit
tags:
- pytorch
- safetensors
- threshold-logic
- neuromorphic
- multiplexer
---

# threshold-mux4

4:1 multiplexer as threshold circuit. Selects one of four data inputs based on 2-bit select signal.

## Circuit

```
d0 ──┐
d1 ──┼──► MUX4 ──► out
d2 ──┤      ▲
d3 ──┘      │
         s1 s0
```

## Truth Table

| s1 | s0 | out |
|----|----|----|
| 0  | 0  | d0 |
| 0  | 1  | d1 |
| 1  | 0  | d2 |
| 1  | 1  | d3 |

## Architecture

| Layer | Neurons | Function |
|-------|---------|----------|
| 1 | 4 | Select gates (one per input) |
| 2 | 1 | OR (combine selections) |

**Total: 5 neurons, 39 parameters, 2 layers**

Each select gate fires only when:
1. Its data input is 1, AND
2. The select lines match its index

## Usage

```python
from safetensors.torch import load_file
import torch

w = load_file('model.safetensors')

def mux4(d0, d1, d2, d3, s1, s0):
    inp = torch.tensor([float(d0), float(d1), float(d2), float(d3),
                        float(s1), float(s0)])
    l1 = (inp @ w['layer1.weight'].T + w['layer1.bias'] >= 0).float()
    out = (l1 @ w['layer2.weight'].T + w['layer2.bias'] >= 0).float()
    return int(out.item())
```

## Files

```
threshold-mux4/
├── model.safetensors
├── create_safetensors.py
├── config.json
└── README.md
```

## License

MIT