{
  "ok": true,
  "data": {
    "entry": {
      "id": "sae",
      "kind": "method",
      "kindLabel": "Method",
      "title": "Sparse autoencoders to untangle neurons",
      "body": "Individual neurons are polysemantic, one neuron fires for 'the Golden Gate Bridge' and also 'feeling trapped' and also semicolons. Train a sparse autoencoder on the activations and you pull those mixed signals apart into cleaner, single-meaning features.",
      "tag": "interpretability",
      "date": "2025 · 09 · 14",
      "publishedAt": "2025-09-14",
      "note": "This is roughly how the 'Golden Gate Claude' demo worked.",
      "insight": {
        "aha": "Sparse features give you named-ish knobs because they separate mixed neuron behavior into cleaner directions.",
        "tryThis": "Treat an SAE feature as a hypothesis, then test whether steering it changes the behavior you care about.",
        "watchFor": [
          "Human-readable labels that overfit a few examples",
          "Feature steering without downstream behavior checks",
          "Polysemantic residue hiding inside a cleaner-looking feature"
        ]
      },
      "extended": [
        "The trick is over-completeness plus sparsity. You map activations into a much wider space but force only a few of those dimensions to be active at once. The pressure pushes tangled concepts apart into separate, more interpretable features.",
        "Once you have the features you can do surgery: amplify one and watch the model fixate on a bridge, suppress another and watch a behavior fade. It is the closest thing we have to named knobs inside the network."
      ],
      "url": "https://tips.platphormnews.com/entries/sae"
    }
  }
}