{
  "ok": true,
  "data": {
    "entry": {
      "id": "activation-patching",
      "kind": "method",
      "kindLabel": "Method",
      "title": "Activation patching to find the cause",
      "body": "Run the model twice, once on a clean prompt and once on a corrupted one, then copy a single activation from one run into the other. If the output flips, you just found a component that carries the behavior. It's causal, not just correlational.",
      "tag": "interpretability",
      "date": "2025 · 07 · 30",
      "publishedAt": "2025-07-30",
      "note": "The closest thing in this field to a controlled experiment.",
      "insight": {
        "aha": "Correlation says a signal is present. Patching asks whether changing that signal changes the outcome.",
        "tryThis": "Patch one layer-position activation from a clean run into a corrupted run and track whether the answer flips.",
        "watchFor": [
          "Large sweeps without a clean causal hypothesis",
          "Output flips caused by formatting rather than the target behavior",
          "A single patch interpreted without replication"
        ]
      },
      "extended": [
        "Probes tell you what is represented. Patching tells you what matters. By swapping one activation and watching the effect, you move from 'this information is present' to 'this component causes the behavior'.",
        "Sweep the patch across layers and positions to localize the circuit. The spots where patching flips the answer are the load-bearing ones; everything else is along for the ride."
      ],
      "url": "https://tips.platphormnews.com/entries/activation-patching"
    }
  }
}