diff --git a/.nojekyll b/.nojekyll
index 3e6c250..3401a52 100644
--- a/.nojekyll
+++ b/.nojekyll
@@ -1 +1 @@
-9f165d84
\ No newline at end of file
+b3fbba8a
\ No newline at end of file
diff --git a/index.html b/index.html
index 67b2365..0af5566 100644
--- a/index.html
+++ b/index.html
@@ -143,7 +143,30 @@
 
 <div class="quarto-listing quarto-listing-container-grid" id="listing-listing">
 <div class="list grid quarto-listing-cols-3">
-<div class="g-col-1" data-index="0" data-listing-date-sort="1687651200000" data-listing-file-modified-sort="1701229733344" data-listing-date-modified-sort="NaN" data-listing-reading-time-sort="7">
+<div class="g-col-1" data-index="0" data-listing-date-sort="1701302400000" data-listing-file-modified-sort="1701380287742" data-listing-date-modified-sort="NaN" data-listing-reading-time-sort="7">
+<a href="./posts/fight_the_illusion.html" class="quarto-grid-link">
+<div class="quarto-grid-item card h-100 card-left">
+<div class="listing-item-img-placeholder card-img-top" style="height: 150px;">&nbsp;</div>
+<div class="card-body post-contents">
+<h5 class="no-anchor card-title listing-title">
+6 Ways to Fight the Interpretability Illusion
+</h5>
+<div class="card-text listing-description">
+Recommended pre-reading:
+</div>
+<div class="card-attribution card-text-small justify">
+<div class="listing-author">
+Michael Sklar
+</div>
+<div class="listing-date">
+Nov 30, 2023
+</div>
+</div>
+</div>
+</div>
+</a>
+</div>
+<div class="g-col-1" data-index="1" data-listing-date-sort="1687651200000" data-listing-file-modified-sort="1701380287738" data-listing-date-modified-sort="NaN" data-listing-reading-time-sort="7">
 <a href="./posts/catalog.html" class="quarto-grid-link">
 <div class="quarto-grid-item card h-100 card-left">
 <p class="card-img-top"><img src="posts/catalog_files/figure-html/cell-9-output-1.png" style="height: 150px;"  class="thumbnail-image card-img"/></p>
diff --git a/listings.json b/listings.json
index 6e75929..3e46da8 100644
--- a/listings.json
+++ b/listings.json
@@ -2,6 +2,7 @@
   {
     "listing": "/index.html",
     "items": [
+      "/posts/fight_the_illusion.html",
       "/posts/catalog.html"
     ]
   }
diff --git a/posts/catalog.html b/posts/catalog.html
index 96df09e..bd1ac14 100644
--- a/posts/catalog.html
+++ b/posts/catalog.html
@@ -798,7 +798,7 @@ <h2 class="anchored" data-anchor-id="github">GitHub</h2>
 });
 </script>
 </div> <!-- /content -->
-<script>var lightboxQuarto = GLightbox({"selector":".lightbox","openEffect":"zoom","descPosition":"bottom","loop":true,"closeEffect":"zoom"});</script>
+<script>var lightboxQuarto = GLightbox({"selector":".lightbox","loop":true,"closeEffect":"zoom","openEffect":"zoom","descPosition":"bottom"});</script>
 
 
 
diff --git a/posts/catalog.out.ipynb b/posts/catalog.out.ipynb
index c6036ef..b4c709b 100644
--- a/posts/catalog.out.ipynb
+++ b/posts/catalog.out.ipynb
@@ -297,7 +297,7 @@
         "Pythia-12B is miscalibrated on 20% of the bigrams and 45% of the\n",
         "trigrams when we ask for prediction of $p \\geq 0.45$."
       ],
-      "id": "b1b12a79-1f9e-4b08-8a4f-de67972b8b34"
+      "id": "f019decb-fb79-4678-b49c-f8ad10685733"
     },
     {
       "cell_type": "code",
@@ -313,7 +313,7 @@
         }
       ],
       "source": [],
-      "id": "77bdcbf8-047e-4347-a779-20da2c9f9647"
+      "id": "361f55f2-44c0-472e-a8b5-c8ca911ab255"
     },
     {
       "cell_type": "markdown",
@@ -375,7 +375,7 @@
         "The dataset is available on Huggingface:\n",
         "[pile_scan_4](https://huggingface.co/datasets/Confirm-Labs/pile_scan_4)"
       ],
-      "id": "5b2c0851-897a-4172-92a6-9d23dbef318d"
+      "id": "49ad1337-e733-40ea-8807-f388dd9d2967"
     },
     {
       "cell_type": "code",
@@ -389,7 +389,7 @@
         }
       ],
       "source": [],
-      "id": "8195339d-751e-456c-bbc0-958d25079862"
+      "id": "3811463e-28a2-4942-8f7d-6d2a31cb6dbb"
     },
     {
       "cell_type": "markdown",
@@ -419,7 +419,7 @@
         "Charles Foster, Jason Phang, et al. 2020. “The Pile: An 800GB Dataset of\n",
         "Diverse Text for Language Modeling.” *arXiv Preprint arXiv:2101.00027*."
       ],
-      "id": "ca8a9289-bbe1-41f9-8569-13f08076feb7"
+      "id": "48252ae7-e749-4523-ad56-d1da5482653b"
     }
   ],
   "nbformat": 4,
diff --git a/posts/fight_the_illusion.html b/posts/fight_the_illusion.html
index 888983d..1bb3151 100644
--- a/posts/fight_the_illusion.html
+++ b/posts/fight_the_illusion.html
@@ -6,7 +6,7 @@
 
 <meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
 
-<meta name="dcterms.date" content="2023-11-28">
+<meta name="dcterms.date" content="2023-11-30">
 
 <title>Confirm - 6 Ways to Fight the Interpretability Illusion</title>
 <style>
@@ -64,10 +64,10 @@
 
 <link rel="stylesheet" href="../styles.css">
 <meta name="citation_title" content="6 Ways to Fight the Interpretability Illusion">
-<meta name="citation_publication_date" content="2023-11-28">
-<meta name="citation_cover_date" content="2023-11-28">
+<meta name="citation_publication_date" content="2023-11-30">
+<meta name="citation_cover_date" content="2023-11-30">
 <meta name="citation_year" content="2023">
-<meta name="citation_online_date" content="2023-11-28">
+<meta name="citation_online_date" content="2023-11-30">
 <meta name="citation_fulltext_html_url" content="https://confirmlabs.org/posts/fight_the_illusion.html">
 <meta name="citation_language" content="en">
 <meta name="citation_reference" content="citation_title=GPT-NeoX-20B: An open-source autoregressive language model;,citation_abstract=We introduce GPT-NeoX-20B, a 20 billion parameter autoregressive language model trained on the Pile, whose weights will be made freely and openly available to the public through a permissive license. It is, to the best of our knowledge, the largest dense autoregressive model that has publicly available weights at the time of submission. In this work, we describe GPT-NeoX-20B’s architecture and training, and evaluate its performance. We open-source the training and evaluation code, as well as the model weights, at https://github.com/EleutherAI/gpt-neox.;,citation_author=Sidney Black;,citation_author=Stella Biderman;,citation_author=Eric Hallahan;,citation_author=Quentin Anthony;,citation_author=Leo Gao;,citation_author=Laurence Golding;,citation_author=Horace He;,citation_author=Connor Leahy;,citation_author=Kyle McDonell;,citation_author=Jason Phang;,citation_author=Michael Pieler;,citation_author=Usvsn Sai Prashanth;,citation_author=Shivanshu Purohit;,citation_author=Laria Reynolds;,citation_author=Jonathan Tow;,citation_author=Ben Wang;,citation_author=Samuel Weinbach;,citation_publication_date=2022-05;,citation_cover_date=2022-05;,citation_year=2022;,citation_fulltext_html_url=https://aclanthology.org/2022.bigscience-1.9;,citation_doi=10.18653/v1/2022.bigscience-1.9;,citation_conference_title=Proceedings of BigScience episode #5 – workshop on challenges &amp;amp;amp; perspectives in creating large language models;,citation_conference=Association for Computational Linguistics;">
@@ -130,7 +130,7 @@ <h1 class="title">6 Ways to Fight the Interpretability Illusion</h1>
     <div>
     <div class="quarto-title-meta-heading">Published</div>
     <div class="quarto-title-meta-contents">
-      <p class="date">November 28, 2023</p>
+      <p class="date">November 30, 2023</p>
     </div>
   </div>
   
@@ -143,11 +143,8 @@ <h1 class="title">6 Ways to Fight the Interpretability Illusion</h1>
 
 <!-----
 
-
-
 Conversion time: 0.504 seconds.
 
-
 Using this Markdown file:
 
 1. Paste this output into your source file.
@@ -163,7 +160,7 @@ <h1 class="title">6 Ways to Fight the Interpretability Illusion</h1>
 ----->
 <p>Recommended pre-reading:</p>
 <ul>
-<li>Atticus Geiger’s <a href="https://arxiv.org/abs/2303.02536">DAS</a> and <a href="https://arxiv.org/pdf/2305.08809.pdf">Boundless DAS</a>.</li>
+<li>Geiger et al.’s <a href="https://arxiv.org/abs/2303.02536">DAS</a> and <a href="https://arxiv.org/pdf/2305.08809.pdf">Boundless DAS</a>.</li>
 <li><a href="https://www.lesswrong.com/posts/RFtkRXHebkwxygDe2/an-interpretability-illusion-for-activation-patching-of">An Interpretability Illusion for Activation Patching of Arbitrary Subspaces</a>.</li>
 <li>The corresponding <a href="https://openreview.net/forum?id=Ebt7JgMHv1">ICLR paper, “Is This the Subspace You Are Looking For?</a>”</li>
 </ul>
@@ -176,7 +173,7 @@ <h1 class="title">6 Ways to Fight the Interpretability Illusion</h1>
 <li>“Pinning” a false belief into the model, for testing or alignment training. For example, forcing the model to believe it is not being watched, in order to test deception or escape behavior.</li>
 </ul>
 The key point is that the interpretability illusion is a failure to <em>describe typical model operation</em>, but a success for <em>enacting the causal model</em>.</li>
-<li><strong>Study more detailed causal models with multiple output streams, multiple options for the input variables, or more compositions.</strong> To start, notice that it is obviously good to have more outputs/consequences of the causal mode in the optimization. Why? First, if we have multiple output-measurements at the end of the causal graph, it is harder for a spurious direction to perform well on all of them by chance. Additionally, if an abstract causal model has modular pieces, then there should be exponentially many combinatorial-swap options that we can test. To score well on the IIA train-loss across all swaps, a spurious structure would have to be very sophisticated. While Lange et al.&nbsp;show that spurious solutions may arise for searches in 1 direction, it should be less likely to occur for <em>pairs</em> of directions, and less likely yet for full spurious circuits. So, illusion problems may be reduced by scaling up model complexity. Some possible issues remain, though:
+<li><strong>Study more detailed causal models with multiple output streams, multiple options for the input variables, or more compositions.</strong> To start, notice that it is obviously good to have more outputs/consequences of the causal mode in the optimization. Why? First, if we have multiple output-measurements at the end of the causal graph, it is harder for a spurious direction to perform well on all of them by chance. Additionally, if an abstract causal model has modular pieces, then there should be exponentially many combinatorial-swap options that we can test. To score well on the optimization’s training-loss across all swaps (in the language of <a href="https://arxiv.org/abs/2303.02536">DAS</a> this is a high “IIA”), the spurious structure would have to be very sophisticated. While Lange et al.&nbsp;show that spurious solutions may arise for searches in 1 direction, it should be less likely to occur for <em>pairs</em> of directions, and less likely yet for full spurious circuits. So, illusion problems may be reduced by scaling up the complexity of the causal model. Some possible issues remain, though:
 <ul>
 <li>In some cases we may struggle to identify specific directions within a multi-part model; i.e., we might find convincing overall performance for a circuit, but an individual dimension or two could be spurious, and we might be unable to determine exactly which.</li>
 <li>This approach relies on big, deep, abstract causal models existing inside the networks, with sufficient robustness in their functioning across variable changes. There is some suggestive work on predictable / standardized structures in LLM’s, from investigations like <a href="https://arxiv.org/pdf/2310.17191.pdf">Feng and Steinhardt (2023</a>)’s entity binding case study, the <a href="https://github.com/redwoodresearch/Easy-Transformer/blob/main/README.md">indirect object identification (IOI)</a> paper, and studies of <a href="https://arxiv.org/pdf/2305.14699.pdf">recursive tasks</a>. However, the consistency/robustness and DAS-discoverability of larger structures in scaled-up models is not yet clear. More case studies in larger models would be valuable.</li>
@@ -187,26 +184,26 @@ <h1 class="title">6 Ways to Fight the Interpretability Illusion</h1>
 <li><strong>Incorporate additional information as a prior / penalty for optimization.</strong> As Lange et al.&nbsp;note in the <a href="https://www.lesswrong.com/posts/RFtkRXHebkwxygDe2/an-interpretability-illusion-for-activation-patching-of">“Illusion” post</a>, and as described in Section 5 of the <a href="https://openreview.net/forum?id=Ebt7JgMHv1">ICLR paper</a>, it is possible to supply additional evidence that a found direction is faithful or not. In the case study with the IOI task, they argued the direction found by DAS on a residual layer fell within the query subspace of human-identified name mover heads. More generally, if intuitions about faithfulness can be scored with a quantitative metric, then tacking that metric onto the optimization as a penalty should help the optimizer favor correct directions over spurious solutions. Still, using this approach requires answering two difficult questions: what additional evidence to choose, and then how to quantify it? Some rough possibilities:
 <ul>
 <li>If we know of structures that should be related to the task, such as entity bindings (<a href="https://arxiv.org/pdf/2310.17191.pdf">Feng and Steinhardt (2023)</a>), we can try to build outwards from them; or if we have a reliable feature dictionary from sparse auto-encoders or “<a href="https://arxiv.org/pdf/2111.13654.pdf">belief graph</a>” per Hase et al.&nbsp;2021 which offers advance predictions for how subsequent layers’ features may react to a change, we can penalize lack of correlation or causal effects on downstream features.</li>
-<li>Somehow draw information from analyzing very basic components of the network: punish “MLP-in-the-middle” solutions by using some combination of changes in MLP activations / attention, gradients, sizes of the induced changes in the residual stream, etc.</li>
+<li>Somehow use the basic structure of the network to quantify which directions are ‘dormant.’ Although this sounds simple, I am unsure how to do it, given the indeterminacy of what a ‘dormant’ direction even means (this issue is described in Lange et al.’s lesswrong <a href="https://www.lesswrong.com/posts/RFtkRXHebkwxygDe2/an-interpretability-illusion-for-activation-patching-of">post</a>, Appendix section: the importance of correct model units.)</li>
 <li>Perhaps next-gen AI will offer accurate “auto-grading”, giving a general yet quantitative evaluation of plausibility of found solutions</li>
 </ul>
-Using extra information in this way unfortunately spends its usability for validation. But if extra information prevents the optimization from getting stuck on false signals, the trade-off should be favorable.</li>
+Using extra information in this way unfortunately spends its usability for validation. But preventing the optimization from getting stuck on spurious signals may be the higher priority.</li>
 </ol>
 <hr>
-<p>Thanks to Atticus Geiger, Jing Huang, Ben Thompson, Zygimantas Straznickas and others for conversations and feedback on earlier drafts.</p>
+<p>Thanks to Atticus Geiger, Jing Huang, Zhengxuan Wu, Ben Thompson, Zygimantas Straznickas and others for conversations and feedback on earlier drafts.</p>
 
 
 
 <div id="quarto-appendix" class="default"><section class="quarto-appendix-contents"><h2 class="anchored quarto-appendix-heading">Citation</h2><div><div class="quarto-appendix-secondary-label">BibTeX citation:</div><pre class="sourceCode code-with-copy quarto-appendix-bibtex"><code class="sourceCode bibtex">@online{sklar2023,
   author = {Sklar, Michael},
   title = {6 {Ways} to {Fight} the {Interpretability} {Illusion}},
-  date = {2023-11-28},
+  date = {2023-11-30},
   url = {https://confirmlabs.org/posts/fight_the_illusion.html},
   langid = {en}
 }
 </code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre><div class="quarto-appendix-secondary-label">For attribution, please cite this work as:</div><div id="ref-sklar2023" class="csl-entry quarto-appendix-citeas" role="listitem">
 Sklar, Michael. 2023. <span>“6 Ways to Fight the Interpretability
-Illusion.”</span> November 28, 2023. <a href="https://confirmlabs.org/posts/fight_the_illusion.html">https://confirmlabs.org/posts/fight_the_illusion.html</a>.
+Illusion.”</span> November 30, 2023. <a href="https://confirmlabs.org/posts/fight_the_illusion.html">https://confirmlabs.org/posts/fight_the_illusion.html</a>.
 </div></div></section></div></main> <!-- /main -->
 <script id="quarto-html-after-body" type="application/javascript">
 window.document.addEventListener("DOMContentLoaded", function (event) {
diff --git a/posts/fight_the_illusion.ipynb b/posts/fight_the_illusion.ipynb
index 2072259..bf02fdd 100644
--- a/posts/fight_the_illusion.ipynb
+++ b/posts/fight_the_illusion.ipynb
@@ -7,9 +7,9 @@
         "# 6 Ways to Fight the Interpretability Illusion\n",
         "\n",
         "Michael Sklar  \n",
-        "2023-11-28"
+        "2023-11-30"
       ],
-      "id": "7861464a-2092-4f6a-b14d-b58ddd31e858"
+      "id": "07417ea9-8e3d-4b06-bbd7-9a019b8b8650"
     },
     {
       "cell_type": "raw",
@@ -19,11 +19,8 @@
       "source": [
         "<!-----\n",
         "\n",
-        "\n",
-        "\n",
         "Conversion time: 0.504 seconds.\n",
         "\n",
-        "\n",
         "Using this Markdown file:\n",
         "\n",
         "1. Paste this output into your source file.\n",
@@ -38,7 +35,7 @@
         "* Source doc: 6 ways to fight the Interpretability illusion\n",
         "----->"
       ],
-      "id": "491d0492-c50e-443c-b7b6-8d60871caad3"
+      "id": "a38da9a3-c507-40c2-91ce-b9aef315812a"
     },
     {
       "cell_type": "markdown",
@@ -46,7 +43,7 @@
       "source": [
         "Recommended pre-reading:\n",
         "\n",
-        "-   Atticus Geiger’s [DAS](https://arxiv.org/abs/2303.02536) and\n",
+        "-   Geiger et al.’s [DAS](https://arxiv.org/abs/2303.02536) and\n",
         "    [Boundless DAS](https://arxiv.org/pdf/2305.08809.pdf).\n",
         "-   [An Interpretability Illusion for Activation Patching of Arbitrary\n",
         "    Subspaces](https://www.lesswrong.com/posts/RFtkRXHebkwxygDe2/an-interpretability-illusion-for-activation-patching-of).\n",
@@ -99,12 +96,14 @@
         "    on all of them by chance. Additionally, if an abstract causal model\n",
         "    has modular pieces, then there should be exponentially many\n",
         "    combinatorial-swap options that we can test. To score well on the\n",
-        "    IIA train-loss across all swaps, a spurious structure would have to\n",
-        "    be very sophisticated. While Lange et al. show that spurious\n",
-        "    solutions may arise for searches in 1 direction, it should be less\n",
-        "    likely to occur for *pairs* of directions, and less likely yet for\n",
-        "    full spurious circuits. So, illusion problems may be reduced by\n",
-        "    scaling up model complexity. Some possible issues remain, though:\n",
+        "    optimization’s training-loss across all swaps (in the language of\n",
+        "    [DAS](https://arxiv.org/abs/2303.02536) this is a high “IIA”), the\n",
+        "    spurious structure would have to be very sophisticated. While Lange\n",
+        "    et al. show that spurious solutions may arise for searches in 1\n",
+        "    direction, it should be less likely to occur for *pairs* of\n",
+        "    directions, and less likely yet for full spurious circuits. So,\n",
+        "    illusion problems may be reduced by scaling up the complexity of the\n",
+        "    causal model. Some possible issues remain, though:\n",
         "    -   In some cases we may struggle to identify specific directions\n",
         "        within a multi-part model; i.e., we might find convincing\n",
         "        overall performance for a circuit, but an individual dimension\n",
@@ -180,26 +179,28 @@
         "        al. 2021 which offers advance predictions for how subsequent\n",
         "        layers’ features may react to a change, we can penalize lack of\n",
         "        correlation or causal effects on downstream features.\n",
-        "    -   Somehow draw information from analyzing very basic components of\n",
-        "        the network: punish “MLP-in-the-middle” solutions by using some\n",
-        "        combination of changes in MLP activations / attention,\n",
-        "        gradients, sizes of the induced changes in the residual stream,\n",
-        "        etc.\n",
+        "    -   Somehow use the basic structure of the network to quantify which\n",
+        "        directions are ‘dormant.’ Although this sounds simple, I am\n",
+        "        unsure how to do it, given the indeterminacy of what a ‘dormant’\n",
+        "        direction even means (this issue is described in Lange et al.’s\n",
+        "        lesswrong\n",
+        "        [post](https://www.lesswrong.com/posts/RFtkRXHebkwxygDe2/an-interpretability-illusion-for-activation-patching-of),\n",
+        "        Appendix section: the importance of correct model units.)\n",
         "    -   Perhaps next-gen AI will offer accurate “auto-grading”, giving a\n",
         "        general yet quantitative evaluation of plausibility of found\n",
         "        solutions\n",
         "\n",
         "    Using extra information in this way unfortunately spends its\n",
-        "    usability for validation. But if extra information prevents the\n",
-        "    optimization from getting stuck on false signals, the trade-off\n",
-        "    should be favorable.\n",
+        "    usability for validation. But preventing the optimization from\n",
+        "    getting stuck on spurious signals may be the higher priority.\n",
         "\n",
         "------------------------------------------------------------------------\n",
         "\n",
-        "Thanks to Atticus Geiger, Jing Huang, Ben Thompson, Zygimantas\n",
-        "Straznickas and others for conversations and feedback on earlier drafts."
+        "Thanks to Atticus Geiger, Jing Huang, Zhengxuan Wu, Ben Thompson,\n",
+        "Zygimantas Straznickas and others for conversations and feedback on\n",
+        "earlier drafts."
       ],
-      "id": "91ffa1d6-7c92-4f86-9a57-f1762355247a"
+      "id": "44919dc4-6ea3-4807-826a-f38f458a67fd"
     }
   ],
   "nbformat": 4,
diff --git a/search.json b/search.json
index 4315e38..4d2713a 100644
--- a/search.json
+++ b/search.json
@@ -1,10 +1,17 @@
 [
+  {
+    "objectID": "posts/fight_the_illusion.html",
+    "href": "posts/fight_the_illusion.html",
+    "title": "6 Ways to Fight the Interpretability Illusion",
+    "section": "",
+    "text": "Recommended pre-reading:\n\nGeiger et al.’s DAS and Boundless DAS.\nAn Interpretability Illusion for Activation Patching of Arbitrary Subspaces.\nThe corresponding ICLR paper, “Is This the Subspace You Are Looking For?”\n\n\nThis post is motivated by Lange, Makelov, and Nanda’s LessWrong post Interpretability Illusion for Activation Patching and ICLR paper. They study Geiger et al’s DAS method, which uses optimization to identify an abstracted causal model with a small subset of dimensions in a neural network’s residual stream or internal MLP layer. Their results show that DAS can, depending on the situation, turn up both “correct” and “spurious” findings on the train-set. From the investigations in the ICLR paper and conversations with a few researchers, my understanding is these “spurious” directions have not performed well on held-out generalization sets, so in practice it is easy to distinguish the “illusions” from “real effects”. But, I am interested in developing even stronger optimize-to-interpret methods. With more powerful optimizers, illusion effects should be even stronger, and competition from spurious signals may make true signals harder to locate in training. So, here are 6 possible ways to fight against the interpretability illusion. Most of them can be tried in combination.\n\nThe causal model still holds, and may still be what we want.: We call it an interpretability illusion because we are failing to describe the model’s normal functioning. But unusual functioning is fine for some goals! Applications include:\n\nFinding latent circuits which might be targetable by optimized non-routine inputs. For example, these circuits might be used in an adversarial attack.\n“Pinning” a false belief into the model, for testing or alignment training. For example, forcing the model to believe it is not being watched, in order to test deception or escape behavior.\n\nThe key point is that the interpretability illusion is a failure to describe typical model operation, but a success for enacting the causal model.\nStudy more detailed causal models with multiple output streams, multiple options for the input variables, or more compositions. To start, notice that it is obviously good to have more outputs/consequences of the causal mode in the optimization. Why? First, if we have multiple output-measurements at the end of the causal graph, it is harder for a spurious direction to perform well on all of them by chance. Additionally, if an abstract causal model has modular pieces, then there should be exponentially many combinatorial-swap options that we can test. To score well on the optimization’s training-loss across all swaps (in the language of DAS this is a high “IIA”), the spurious structure would have to be very sophisticated. While Lange et al. show that spurious solutions may arise for searches in 1 direction, it should be less likely to occur for pairs of directions, and less likely yet for full spurious circuits. So, illusion problems may be reduced by scaling up the complexity of the causal model. Some possible issues remain, though:\n\nIn some cases we may struggle to identify specific directions within a multi-part model; i.e., we might find convincing overall performance for a circuit, but an individual dimension or two could be spurious, and we might be unable to determine exactly which.\nThis approach relies on big, deep, abstract causal models existing inside the networks, with sufficient robustness in their functioning across variable changes. There is some suggestive work on predictable / standardized structures in LLM’s, from investigations like Feng and Steinhardt (2023)’s entity binding case study, the indirect object identification (IOI) paper, and studies of recursive tasks. However, the consistency/robustness and DAS-discoverability of larger structures in scaled-up models is not yet clear. More case studies in larger models would be valuable.\n\nMeasure generalizability, and use it to filter out spurious findings after-the-fact. This is just common-sense, and researchers are already doing this in several ways. We can construct train/test splits with random sampling, and conclude a found direction is spurious if it does not generalize on the test data; or we could ask how the patched model generalizes out-of-training-distribution following a small perturbation, such as adding extra preceding tokens. Spurious solutions are likely to be sensitive to minor changes, and for many purposes we are primarily interested in causal models that generalize well. As mentioned earlier, the ICLR paper’s `spurious’ findings performed sufficiently poorly on generalization sets that they could easily be distinguished from real effects.\nQuantify a null distribution. In the “Illusion” post, Lange et al. show that the strength of the spurious signal depends on how many neurons it is allowed to optimize over. So, a very strong signal, taken over a small optimization set, should be more convincing. Thinking as statisticians, we could attempt to construct a null distribution for the spurious signals; this approach could offer evidence that a causal map element is being represented at all. We can do this inference for individual pieces of a larger causal model, with each component having its own uncertainty.\nUse unsupervised feature extraction as a first step. Recent interpretability work with auto-encoders (Bricken et al. 2023, Cunningham et al. 2023) suggests that many of a small transformer’s most important features can be identified. If this technique scales well, it could vastly reduce the amount of optimization pressure needed to identify the right directions, shrinking the search space and reducing optimistic bias and spurious findings.\nIncorporate additional information as a prior / penalty for optimization. As Lange et al. note in the “Illusion” post, and as described in Section 5 of the ICLR paper, it is possible to supply additional evidence that a found direction is faithful or not. In the case study with the IOI task, they argued the direction found by DAS on a residual layer fell within the query subspace of human-identified name mover heads. More generally, if intuitions about faithfulness can be scored with a quantitative metric, then tacking that metric onto the optimization as a penalty should help the optimizer favor correct directions over spurious solutions. Still, using this approach requires answering two difficult questions: what additional evidence to choose, and then how to quantify it? Some rough possibilities:\n\nIf we know of structures that should be related to the task, such as entity bindings (Feng and Steinhardt (2023)), we can try to build outwards from them; or if we have a reliable feature dictionary from sparse auto-encoders or “belief graph” per Hase et al. 2021 which offers advance predictions for how subsequent layers’ features may react to a change, we can penalize lack of correlation or causal effects on downstream features.\nSomehow use the basic structure of the network to quantify which directions are ‘dormant.’ Although this sounds simple, I am unsure how to do it, given the indeterminacy of what a ‘dormant’ direction even means (this issue is described in Lange et al.’s lesswrong post, Appendix section: the importance of correct model units.)\nPerhaps next-gen AI will offer accurate “auto-grading”, giving a general yet quantitative evaluation of plausibility of found solutions\n\nUsing extra information in this way unfortunately spends its usability for validation. But preventing the optimization from getting stuck on spurious signals may be the higher priority.\n\n\nThanks to Atticus Geiger, Jing Huang, Zhengxuan Wu, Ben Thompson, Zygimantas Straznickas and others for conversations and feedback on earlier drafts.\n\n\n\nCitationBibTeX citation:@online{sklar2023,\n  author = {Sklar, Michael},\n  title = {6 {Ways} to {Fight} the {Interpretability} {Illusion}},\n  date = {2023-11-30},\n  url = {https://confirmlabs.org/posts/fight_the_illusion.html},\n  langid = {en}\n}\nFor attribution, please cite this work as:\nSklar, Michael. 2023. “6 Ways to Fight the Interpretability\nIllusion.” November 30, 2023. https://confirmlabs.org/posts/fight_the_illusion.html."
+  },
   {
     "objectID": "index.html",
     "href": "index.html",
     "title": "Confirm",
     "section": "",
-    "text": "A catalog of several million tasks Pythia can do.\n\n\n\n\n\n\nT. Ben Thompson, Michael Sklar\n\n\nJun 25, 2023\n\n\n\n\n\n\n\n\nNo matching items"
+    "text": "6 Ways to Fight the Interpretability Illusion\n\n\n\n\n\n\nMichael Sklar\n\n\nNov 30, 2023\n\n\n\n\n\n\n\n\n\n\n\n\nA catalog of several million tasks Pythia can do.\n\n\n\n\n\n\nT. Ben Thompson, Michael Sklar\n\n\nJun 25, 2023\n\n\n\n\n\n\n\n\nNo matching items"
   },
   {
     "objectID": "posts/catalog.html",
diff --git a/sitemap.xml b/sitemap.xml
index e22a1bd..1b1d5ff 100644
--- a/sitemap.xml
+++ b/sitemap.xml
@@ -1,11 +1,15 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+  <url>
+    <loc>https://confirmlabs.org/posts/fight_the_illusion.html</loc>
+    <lastmod>2023-11-30T21:38:21.542Z</lastmod>
+  </url>
   <url>
     <loc>https://confirmlabs.org/index.html</loc>
-    <lastmod>2023-11-29T03:49:03.396Z</lastmod>
+    <lastmod>2023-11-30T21:38:17.678Z</lastmod>
   </url>
   <url>
     <loc>https://confirmlabs.org/posts/catalog.html</loc>
-    <lastmod>2023-11-29T03:49:06.212Z</lastmod>
+    <lastmod>2023-11-30T21:38:20.458Z</lastmod>
   </url>
 </urlset>