4-bit bitwise NOT, magnitude 4

README.md

@@ -0,0 +1,71 @@
+---
+license: mit
+tags:
+- pytorch
+- safetensors
+- threshold-logic
+- neuromorphic
+---
+
+# threshold-negator4bit
+
+4-bit bitwise NOT (one's complement negation).
+
+## Function
+
+negator4bit(a3, a2, a1, a0) = [NOT(a3), NOT(a2), NOT(a1), NOT(a0)]
+
+Inverts each bit independently.
+
+## Truth Table (selected rows)
+
+| Input | Output |
+|-------|--------|
+| 0000 | 1111 |
+| 0001 | 1110 |
+| 0101 | 1010 |
+| 1111 | 0000 |
+
+## Architecture
+
+Single layer with 4 independent NOT neurons.
+
+| Output | Weight on input | Bias |
+|--------|-----------------|------|
+| y3 | a3: -1 | 0 |
+| y2 | a2: -1 | 0 |
+| y1 | a1: -1 | 0 |
+| y0 | a0: -1 | 0 |
+
+Each neuron fires when its input is 0.
+
+## Parameters
+
+| | |
+|---|---|
+| Inputs | 4 |
+| Outputs | 4 |
+| Neurons | 4 |
+| Layers | 1 |
+| Parameters | 8 |
+| Magnitude | 4 |
+
+## Usage
+
+```python
+from safetensors.torch import load_file
+import torch
+
+w = load_file('model.safetensors')
+
+def negator4(a3, a2, a1, a0):
+    inp = torch.tensor([float(a3), float(a2), float(a1), float(a0)])
+    return [int((inp * w[f'y{i}.weight']).sum() + w[f'y{i}.bias'] >= 0)
+            for i in range(4)]
+
+print(negator4(0, 1, 0, 1))  # [1, 0, 1, 0]
+```
+
+## License
+
+MIT

config.json

@@ -0,0 +1,9 @@
+{
+  "name": "threshold-negator4bit",
+  "description": "4-bit bitwise NOT (one's complement)",
+  "inputs": 4,
+  "outputs": 4,
+  "neurons": 4,
+  "layers": 1,
+  "parameters": 8
+}

create_safetensors.py

@@ -0,0 +1,41 @@
+import torch
+from safetensors.torch import save_file
+
+weights = {}
+
+# Each output yi = NOT(ai)
+# NOT(x): weight -1, bias 0, fires when x = 0
+for i in range(4):
+    w = [0.0, 0.0, 0.0, 0.0]
+    w[i] = -1.0
+    weights[f'y{i}.weight'] = torch.tensor([w], dtype=torch.float32)
+    weights[f'y{i}.bias'] = torch.tensor([0.0], dtype=torch.float32)
+
+save_file(weights, 'model.safetensors')
+
+# Verify
+def negator4(a3, a2, a1, a0):
+    inp = torch.tensor([float(a3), float(a2), float(a1), float(a0)])
+    outputs = []
+    for i in range(4):
+        y = int((inp * weights[f'y{i}.weight']).sum() + weights[f'y{i}.bias'] >= 0)
+        outputs.append(y)
+    return outputs
+
+print("Verifying negator4bit...")
+errors = 0
+for i in range(16):
+    a3, a2, a1, a0 = (i >> 3) & 1, (i >> 2) & 1, (i >> 1) & 1, i & 1
+    result = negator4(a3, a2, a1, a0)
+    expected = [1 - a3, 1 - a2, 1 - a1, 1 - a0]
+    if result != expected:
+        errors += 1
+        print(f"ERROR: {a3}{a2}{a1}{a0} -> {result}, expected {expected}")
+
+if errors == 0:
+    print("All 16 test cases passed!")
+else:
+    print(f"FAILED: {errors} errors")
+
+mag = sum(t.abs().sum().item() for t in weights.values())
+print(f"Magnitude: {mag:.0f}")

model.py

@@ -0,0 +1,24 @@
+import torch
+from safetensors.torch import load_file
+
+def load_model(path='model.safetensors'):
+    return load_file(path)
+
+def negator4(a3, a2, a1, a0, weights):
+    """Bitwise NOT of 4-bit input (one's complement)."""
+    inp = torch.tensor([float(a3), float(a2), float(a1), float(a0)])
+    outputs = []
+    for i in range(4):
+        y = int((inp * weights[f'y{i}.weight']).sum() + weights[f'y{i}.bias'] >= 0)
+        outputs.append(y)
+    return outputs
+
+if __name__ == '__main__':
+    w = load_model()
+    print('negator4bit truth table:')
+    print('input | output')
+    print('------+-------')
+    for i in range(16):
+        a3, a2, a1, a0 = (i >> 3) & 1, (i >> 2) & 1, (i >> 1) & 1, i & 1
+        result = negator4(a3, a2, a1, a0, w)
+        print(f'{a3}{a2}{a1}{a0}  | {"".join(map(str, result))}')