{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreieysafmzivde36tbcfpo32y5tjz7eoigrks5f2i6y24umvdhjgyhi",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mgroob2s43h2"
  },
  "path": "/t/experimental-protocol-proposal-testing-the-prompt-coherence-engine-pce/174041#post_6",
  "publishedAt": "2026-03-11T08:49:21.000Z",
  "site": "https://discuss.huggingface.co",
  "textContent": "Hi Allan, thanks for the thoughtful response.\n\nYour point about the PCE targeting a specific behavioral signature — stability under contradictory constraints — rather than just higher consistency in general is a useful distinction. That would actually change the prediction for the hidden-state signal: instead of uniformly high coherence, you’d expect a particular trajectory shape — maybe a spike followed by stabilization, as the model navigates the contradiction and settles.\n\nThe suggestion to apply hidden-state measurements to the PCE condition is something I’d genuinely be interested in. The most direct test would be: does the coherence trajectory look different under Condition C (PCE) vs. Condition B (long neutral baseline) for the same dilemma? If the PCE is doing what the hypothesis predicts, the internal dynamics should diverge even when surface outputs look similar.\n\nOne practical note: the signal I’m using (cosine similarity at Layer 27) is most informative in iterative settings where the model builds on its own prior outputs over multiple steps. For single-turn dilemma responses it might need adjustment — either a different layer, or tracking similarity across multiple sampled completions rather than across generation steps.\n\nIf you run the protocol and want to add hidden-state logging as an optional arm, I’m happy to share the hook implementation. It’s a straightforward forward hook, about 30 lines.",
  "title": "Experimental Protocol Proposal: Testing the Prompt Coherence Engine (PCE)"
}