File size: 1,192 Bytes
ccee2d3 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 | ---
license: mit
tags:
- pytorch
- safetensors
- threshold-logic
- neuromorphic
---
# threshold-signbit4
Extract sign bit (MSB) of 4-bit number.
## Function
signbit4(a3, a2, a1, a0) = a3
In 2's complement representation:
- Output 0: non-negative (0 to 7)
- Output 1: negative (-8 to -1)
## Truth Table
| a3 | a2 | a1 | a0 | unsigned | signed | signbit |
|----|----|----|----|---------:|-------:|:-------:|
| 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| 0 | 1 | 1 | 1 | 7 | +7 | 0 |
| 1 | 0 | 0 | 0 | 8 | -8 | 1 |
| 1 | 1 | 1 | 1 | 15 | -1 | 1 |
## Architecture
Single neuron extracting MSB:
- Weights: [1, 0, 0, 0]
- Bias: -1
Fires when: a3 - 1 >= 0, i.e., a3 >= 1
## Parameters
| | |
|---|---|
| Inputs | 4 |
| Outputs | 1 |
| Neurons | 1 |
| Layers | 1 |
| Parameters | 5 |
| Magnitude | 2 |
## Usage
```python
from safetensors.torch import load_file
import torch
w = load_file('model.safetensors')
def signbit4(a3, a2, a1, a0):
inp = torch.tensor([float(a3), float(a2), float(a1), float(a0)])
return int((inp @ w['neuron.weight'].T + w['neuron.bias'] >= 0).item())
print(signbit4(0, 1, 1, 1)) # 0 (value = +7)
print(signbit4(1, 0, 0, 0)) # 1 (value = -8)
```
## License
MIT
|