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

# threshold-5outof8

Strict majority detector. Fires when more than half (5+ of 8) inputs are set.

## Circuit

```
  x₀ x₁ x₂ x₃ x₄ x₅ x₆ x₇
   │  │  │  │  │  │  │  │
   └──┴──┴──┴──┼──┴──┴──┴──┘
               ▼
          ┌─────────┐
          │ w: all 1│
          │ b:  -5  │
          └─────────┘
               │
               ▼
            HW ≥ 5
```

## Mechanism

- Sum = (number of 1s) - 5
- Fires when Hamming weight ≥ 5

This is the majority threshold for 8 inputs. A tie (4-4) doesn't pass. You need more 1s than 0s.

Functionally equivalent to threshold-majority.

## k-out-of-8 Family

| Circuit | Bias | Fires when |
|---------|------|------------|
| ... | ... | ... |
| 4-out-of-8 | -4 | HW ≥ 4 |
| **5-out-of-8** | **-5** | **HW ≥ 5 (this = majority)** |
| 6-out-of-8 | -6 | HW ≥ 6 |
| ... | ... | ... |

## Parameters

| | |
|---|---|
| Weights | [1, 1, 1, 1, 1, 1, 1, 1] |
| Bias | -5 |
| Total | 9 parameters |

## Usage

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

w = load_file('model.safetensors')

def majority(bits):
    inputs = torch.tensor([float(b) for b in bits])
    return int((inputs * w['weight']).sum() + w['bias'] >= 0)
```

## Files

```
threshold-5outof8/
├── model.safetensors
├── model.py
├── config.json
└── README.md
```

## License

MIT