Skip to content

Commit

Permalink
updated news
Browse files Browse the repository at this point in the history
  • Loading branch information
pat-alt committed Aug 29, 2024
1 parent 89774d2 commit f22362a
Show file tree
Hide file tree
Showing 41 changed files with 2,592 additions and 2,614 deletions.
14 changes: 0 additions & 14 deletions blog/posts/empirical-bayes/index.qmd

This file was deleted.

10 changes: 4 additions & 6 deletions content/news.qmd
Original file line number Diff line number Diff line change
@@ -1,11 +1,9 @@
### News

- In May 2024, I'll be in London, UK, to present research at [ECONDAT 2024](/content/talks/posts/2024-econdat/) (King's College London), [The Alan Turing Institute](/content/talks/posts/2024-turing/) and [Imperial College London](/content/talks/posts/2024-imperial/). If you're around, come say hi!
- Our position paper "Against Spurious Sparks---Dovelating Inflated AI Claims 🕊️" has been accepted at ICML 2024. We argue that finding patterns in latent embeddings of models like LLMs is not surprising and call for the academic community to exercise extra caution when interpreting such findings. [[preprint](https://arxiv.org/abs/2402.03962)], [[blog post](/blog/posts/spurious-sparks/index.qmd)]
- I was at [AAAI 2024](https://aaai.org/aaai-conference/) in Vancouver, Canada presenting our work on *Faithful Model Explanations through Energy-Constrained Conformal Counterfactuals*. If you're around, come say hi! [[preprint](https://arxiv.org/abs/2312.10648)], [[blog](/blog/posts/eccco/index.qmd)], [[TDS](https://medium.com/towards-data-science/ecccos-from-the-black-box-c4bd6ef20263)], [[code](https://github.com/pat-alt/ECCCo.jl)], [[slides/poster](/content/talks/posts/2024-aaai/index.qmd)]
- I've published a new lightweight Julia package for working with the Trillion Dollar Words dataset, model and embeddings. Find out more about it [here](/blog/posts/trillion-dollar-words/index.qmd).
- The [Taija](https://github.com/JuliaTrustworthyAI) package ecosystem for Trustworthy AI in Julia is slowly growing. If you're interested in contributing, please get in touch!
- As of September 15, 2023, I'm a third-year PhD student. I'm still loving the experience, although second year did get a bit stressful at times. I'm looking forward to the next two years!
- Taija, the organization that hosts software geared towards Trustworthy Artificial Intelligence in Julia, has its own website now: [www.taija.org](https://www.taija.org/). The ecosystem keeps slowly growing. If you're interested in contributing, please get in touch!
- In July 2024, I presented our position paper "Stop Making Unscientific AGI Performance Claims" at ICML 2024. We argue that finding patterns in latent embeddings of models like LLMs is not surprising and call for the academic community to exercise extra caution when interpreting such findings. [[preprint](https://arxiv.org/abs/2402.03962)], [[blog post](/blog/posts/spurious-sparks/index.qmd)], [[slides/poster](/content/talks/posts/2024-icml/index.qmd)]
- In May 2024, I was in London to present research at [ECONDAT 2024](/content/talks/posts/2024-econdat/) (King's College London), [The Alan Turing Institute](/content/talks/posts/2024-turing/) and [Imperial College London](/content/talks/posts/2024-imperial/).
- In Frebruary 2024, I attended [AAAI 2024](https://aaai.org/aaai-conference/) in Vancouver, Canada presenting our work on *Faithful Model Explanations through Energy-Constrained Conformal Counterfactuals*. If you're around, come say hi! [[preprint](https://arxiv.org/abs/2312.10648)], [[blog](/blog/posts/eccco/index.qmd)], [[TDS](https://medium.com/towards-data-science/ecccos-from-the-black-box-c4bd6ef20263)], [[code](https://github.com/pat-alt/ECCCo.jl)], [[slides/poster](/content/talks/posts/2024-aaai/index.qmd)]

### Featured Visual

Expand Down
2 changes: 1 addition & 1 deletion content/publications/posts/icml-2024/index.qmd
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
---
title: "Position Paper: Against Spurious Sparks − Dovelating Inflated AI Claims 🕊️"
title: "Position: Stop Making Unscientific AGI Performance Claims"
date: '2024-05-07'
venue: Forty-first International Conference on Machine Learning (ICML)
categories:
Expand Down
2 changes: 0 additions & 2 deletions docs/blog/posts/empirical-bayes/index.html

This file was deleted.

Binary file modified docs/content/about/cv_compact.pdf
Binary file not shown.
10 changes: 4 additions & 6 deletions docs/content/news.html
Original file line number Diff line number Diff line change
Expand Up @@ -174,12 +174,10 @@ <h2 id="toc-title">On this page</h2>
<section id="news" class="level3">
<h3 class="anchored" data-anchor-id="news">News</h3>
<ul>
<li>In May 2024, I’ll be in London, UK, to present research at <a href="../content/talks/posts/2024-econdat/">ECONDAT 2024</a> (King’s College London), <a href="../content/talks/posts/2024-turing/">The Alan Turing Institute</a> and <a href="../content/talks/posts/2024-imperial/">Imperial College London</a>. If you’re around, come say hi!</li>
<li>Our position paper “Against Spurious Sparks—Dovelating Inflated AI Claims 🕊️” has been accepted at ICML 2024. We argue that finding patterns in latent embeddings of models like LLMs is not surprising and call for the academic community to exercise extra caution when interpreting such findings. [<a href="https://arxiv.org/abs/2402.03962">preprint</a>], [<a href="../blog/posts/spurious-sparks/index.html">blog post</a>]</li>
<li>I was at <a href="https://aaai.org/aaai-conference/">AAAI 2024</a> in Vancouver, Canada presenting our work on <em>Faithful Model Explanations through Energy-Constrained Conformal Counterfactuals</em>. If you’re around, come say hi! [<a href="https://arxiv.org/abs/2312.10648">preprint</a>], [<a href="../blog/posts/eccco/index.html">blog</a>], [<a href="https://medium.com/towards-data-science/ecccos-from-the-black-box-c4bd6ef20263">TDS</a>], [<a href="https://github.com/pat-alt/ECCCo.jl">code</a>], [<a href="../content/talks/posts/2024-aaai/index.html">slides/poster</a>]</li>
<li>I’ve published a new lightweight Julia package for working with the Trillion Dollar Words dataset, model and embeddings. Find out more about it <a href="../blog/posts/trillion-dollar-words/index.html">here</a>.</li>
<li>The <a href="https://github.com/JuliaTrustworthyAI">Taija</a> package ecosystem for Trustworthy AI in Julia is slowly growing. If you’re interested in contributing, please get in touch!</li>
<li>As of September 15, 2023, I’m a third-year PhD student. I’m still loving the experience, although second year did get a bit stressful at times. I’m looking forward to the next two years!</li>
<li>Taija, the organization that hosts software geared towards Trustworthy Artificial Intelligence in Julia, has its own website now: <a href="https://www.taija.org/">www.taija.org</a>. The ecosystem keeps slowly growing. If you’re interested in contributing, please get in touch!</li>
<li>In July 2024, I presented our position paper “Stop Making Unscientific AGI Performance Claims” at ICML 2024. We argue that finding patterns in latent embeddings of models like LLMs is not surprising and call for the academic community to exercise extra caution when interpreting such findings. [<a href="https://arxiv.org/abs/2402.03962">preprint</a>], [<a href="../blog/posts/spurious-sparks/index.html">blog post</a>], [<a href="../content/talks/posts/2024-icml/index.html">slides/poster</a>]</li>
<li>In May 2024, I was in London to present research at <a href="../content/talks/posts/2024-econdat/">ECONDAT 2024</a> (King’s College London), <a href="../content/talks/posts/2024-turing/">The Alan Turing Institute</a> and <a href="../content/talks/posts/2024-imperial/">Imperial College London</a>.</li>
<li>In Frebruary 2024, I attended <a href="https://aaai.org/aaai-conference/">AAAI 2024</a> in Vancouver, Canada presenting our work on <em>Faithful Model Explanations through Energy-Constrained Conformal Counterfactuals</em>. If you’re around, come say hi! [<a href="https://arxiv.org/abs/2312.10648">preprint</a>], [<a href="../blog/posts/eccco/index.html">blog</a>], [<a href="https://medium.com/towards-data-science/ecccos-from-the-black-box-c4bd6ef20263">TDS</a>], [<a href="https://github.com/pat-alt/ECCCo.jl">code</a>], [<a href="../content/talks/posts/2024-aaai/index.html">slides/poster</a>]</li>
</ul>
</section>
<section id="featured-visual" class="level3">
Expand Down
4 changes: 2 additions & 2 deletions docs/content/publications/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -287,9 +287,9 @@ <h1 class="title">Publications</h1>
</tr>
</thead>
<tbody class="list">
<tr data-index="0" data-categories="mechanistic interpretability,large language models,artificial intelligence,anthropomorphism" data-listing-date-sort="1715032800000" data-listing-file-modified-sort="1715249081225" data-listing-date-modified-sort="NaN" data-listing-reading-time-sort="2" data-listing-word-count-sort="267" data-listing-title-sort="Position Paper: Against Spurious Sparks − Dovelating Inflated AI Claims 🕊️" data-listing-filename-sort="index.qmd">
<tr data-index="0" data-categories="mechanistic interpretability,large language models,artificial intelligence,anthropomorphism" data-listing-date-sort="1715032800000" data-listing-file-modified-sort="1724943258788" data-listing-date-modified-sort="NaN" data-listing-reading-time-sort="2" data-listing-word-count-sort="267" data-listing-title-sort="Position: Stop Making Unscientific AGI Performance Claims" data-listing-filename-sort="index.qmd">
<td>
<a href="../../content/publications/posts/icml-2024/index.html" class="title listing-title">Position Paper: Against Spurious Sparks − Dovelating Inflated AI Claims 🕊️</a>
<a href="../../content/publications/posts/icml-2024/index.html" class="title listing-title">Position: Stop Making Unscientific AGI Performance Claims</a>
</td>
<td>
<span class="listing-date">2024</span>
Expand Down
6 changes: 3 additions & 3 deletions docs/content/publications/posts/icml-2024/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@
<meta name="dcterms.date" content="2024-05-07">
<meta name="description" content="We call for the academic community to exercise caution in interpreting and communicating about AI research outcomes.">

<title>Position Paper: Against Spurious Sparks − Dovelating Inflated AI Claims 🕊️ – patalt</title>
<title>Position: Stop Making Unscientific AGI Performance Claims – patalt</title>
<style>
code{white-space: pre-wrap;}
span.smallcaps{font-variant: small-caps;}
Expand Down Expand Up @@ -81,7 +81,7 @@


<link rel="stylesheet" href="../../../../styles.css">
<meta property="og:title" content="Position Paper: Against Spurious Sparks − Dovelating Inflated AI Claims 🕊️ – patalt">
<meta property="og:title" content="Position: Stop Making Unscientific AGI Performance Claims – patalt">
<meta property="og:description" content="We call for the academic community to exercise caution in interpreting and communicating about AI research outcomes.">
<meta property="og:site_name" content="patalt">
</head>
Expand Down Expand Up @@ -172,7 +172,7 @@ <h2 id="toc-title">On this page</h2>

<header id="title-block-header" class="quarto-title-block default">
<div class="quarto-title">
<h1 class="title">Position Paper: Against Spurious Sparks − Dovelating Inflated AI Claims 🕊️</h1>
<h1 class="title">Position: Stop Making Unscientific AGI Performance Claims</h1>
<div class="quarto-categories">
<div class="quarto-category">mechanistic interpretability</div>
<div class="quarto-category">large language models</div>
Expand Down
4 changes: 2 additions & 2 deletions docs/content/talks/posts/2022-boe/presentation.html
Original file line number Diff line number Diff line change
Expand Up @@ -781,15 +781,15 @@
Q(95);Q(66);Q(64);Q(65);
Q(76);
Q(108);Q(105);Q(77);
Q(5);Q(67);Q(18);Q(19);
Q(5);Q(67);
Q(90);lb[1]=ei('1',6E4);
lb[3]=ei('10',1);lb[2]=ei('',50);Q(24);Q(12);
Q(63);Q(96);
Q(8);
Q(52);
Q(23);Q(42);
Q(80);Q(93);Q(94);Q(91);Q(82);Q(49);Q(51);Q(61);Q(13);Q(48);
Q(68);function T(a){return!!fi[a]}var hi={},ii=E.google_tag_manager=E.google_tag_manager||{};hi.Rg="48r0";hi.oe=Number("0")||0;hi.hb="dataLayer";hi.fn="ChAI8LjAtgYQ8/fl7L7C6eFxEiUA4dkr/R1VbT1B8IHUWeDt4h1eG26vxYe1InzE6B2gWwCvO0GtGgKNbQ\x3d\x3d";var ji={__cl:1,__ecl:1,__ehl:1,__evl:1,__fal:1,__fil:1,__fsl:1,__hl:1,__jel:1,__lcl:1,__sdl:1,__tl:1,__ytl:1},ki={__paused:1,__tg:1},li;for(li in ji)ji.hasOwnProperty(li)&&(ki[li]=1);var mi=xb(""),ni=!1,oi,pi=!1;pi=!0;
Q(68);function T(a){return!!fi[a]}var hi={},ii=E.google_tag_manager=E.google_tag_manager||{};hi.Rg="48r0";hi.oe=Number("0")||0;hi.hb="dataLayer";hi.fn="ChAI8LjAtgYQ8/fl7L7C6eFxEiUA4dkr/e37Svb6Mm+M1kNK4mbDWFKtf0Sl4NdG5yLEylsQ58OCGgI9lA\x3d\x3d";var ji={__cl:1,__ecl:1,__ehl:1,__evl:1,__fal:1,__fil:1,__fsl:1,__hl:1,__jel:1,__lcl:1,__sdl:1,__tl:1,__ytl:1},ki={__paused:1,__tg:1},li;for(li in ji)ji.hasOwnProperty(li)&&(ki[li]=1);var mi=xb(""),ni=!1,oi,pi=!1;pi=!0;
oi=pi;var qi,ri=!1;qi=ri;var si,ti=!1;si=ti;hi.Cd="www.googletagmanager.com";var ui=""+hi.Cd+(oi?"/gtag/js":"/gtm.js"),vi=null,wi=null,xi={},yi={};function zi(){var a=ii.sequence||1;ii.sequence=a+1;return a}hi.mk="true";var Ai="";hi.Sg=Ai;var Bi=new function(){this.j="";this.H=!1;this.D=0;this.R=this.vb=this.Z=this.Pa=this.M=""};function Ci(){var a=Bi.M.length;return Bi.M[a-1]==="/"?Bi.M.substring(0,a-1):Bi.M}function Di(){return Bi.H&&Bi.D!==1}function Ei(a){for(var b={},c=na(a.split("|")),d=c.next();!d.done;d=c.next())b[d.value]=!0;return b}var Fi=new ub,Gi={},Hi={},Ki={name:hi.hb,set:function(a,b){h(Jb(a,b),Gi);Ii()},get:function(a){return Ji(a,2)},reset:function(){Fi=new ub;Gi={};Ii()}};function Ji(a,b){return b!=2?Fi.get(a):Li(a)}function Li(a,b){var c=a.split(".");b=b||[];for(var d=Gi,e=0;e<c.length;e++){if(d===null)return!1;if(d===void 0)break;d=d[c[e]];if(b.indexOf(d)!==-1)return}return d}function Mi(a,b){Hi.hasOwnProperty(a)||(Fi.set(a,b),h(Jb(a,b),Gi),Ii())}
function Ni(){for(var a=["gtm.allowlist","gtm.blocklist","gtm.whitelist","gtm.blacklist","tagTypeBlacklist"],b=0;b<a.length;b++){var c=a[b],d=Ji(c,1);if(Array.isArray(d)||Ta(d))d=h(d);Hi[c]=d}}function Ii(a){z(Hi,function(b,c){Fi.set(b,c);h(Jb(b),Gi);h(Jb(b,c),Gi);a&&delete Hi[b]})}function Oi(a,b){var c,d=(b===void 0?2:b)!==1?Li(a):Fi.get(a);Qa(d)==="array"||Qa(d)==="object"?c=h(d):c=d;return c};
var Pi=function(a,b,c){if(!c)return!1;var d=c.selector_type,e=String(c.value),f;if(d==="js_variable"){e=e.replace(/\["?'?/g,".").replace(/"?'?\]/g,"");for(var g=e.split(","),k=0;k<g.length;k++){var m=g[k].trim();if(m){if(Gb(m,"dataLayer."))f=Ji(m.substring(10));else{var n=m.split(".");f=E[n.shift()];for(var p=0;p<n.length;p++)f=f&&f[n[p]]}if(f!==void 0)break}}}else if(d==="css_selector"&&lh)try{var q=kh(e);if(q&&q.length>0){f=[];for(var r=0;r<q.length&&r<(b==="email"||b==="phone_number"?5:1);r++)f.push(Fc(q[r])||
Expand Down
4 changes: 2 additions & 2 deletions docs/content/talks/posts/2022-delft-fintech/presentation.html
Original file line number Diff line number Diff line change
Expand Up @@ -781,15 +781,15 @@
Q(95);Q(66);Q(64);Q(65);
Q(76);
Q(108);Q(105);Q(77);
Q(5);Q(67);
Q(5);Q(67);Q(18);
Q(90);lb[1]=ei('1',6E4);
lb[3]=ei('10',1);lb[2]=ei('',50);Q(24);Q(12);
Q(63);Q(96);
Q(8);
Q(52);
Q(23);Q(42);
Q(80);Q(93);Q(94);Q(91);Q(82);Q(49);Q(51);Q(61);Q(13);Q(48);
Q(68);function T(a){return!!fi[a]}var hi={},ii=E.google_tag_manager=E.google_tag_manager||{};hi.Rg="48r0";hi.oe=Number("0")||0;hi.hb="dataLayer";hi.fn="ChAI8LjAtgYQ8/fl7L7C6eFxEiUA4dkr/R1VbT1B8IHUWeDt4h1eG26vxYe1InzE6B2gWwCvO0GtGgKNbQ\x3d\x3d";var ji={__cl:1,__ecl:1,__ehl:1,__evl:1,__fal:1,__fil:1,__fsl:1,__hl:1,__jel:1,__lcl:1,__sdl:1,__tl:1,__ytl:1},ki={__paused:1,__tg:1},li;for(li in ji)ji.hasOwnProperty(li)&&(ki[li]=1);var mi=xb(""),ni=!1,oi,pi=!1;pi=!0;
Q(68);function T(a){return!!fi[a]}var hi={},ii=E.google_tag_manager=E.google_tag_manager||{};hi.Rg="48r0";hi.oe=Number("0")||0;hi.hb="dataLayer";hi.fn="ChAI8LjAtgYQ8/fl7L7C6eFxEiUA4dkr/b9Z7jZaOFNpZVCOeQFc31BXkyRM6/WW7GkKZVSqEljbGgJY2A\x3d\x3d";var ji={__cl:1,__ecl:1,__ehl:1,__evl:1,__fal:1,__fil:1,__fsl:1,__hl:1,__jel:1,__lcl:1,__sdl:1,__tl:1,__ytl:1},ki={__paused:1,__tg:1},li;for(li in ji)ji.hasOwnProperty(li)&&(ki[li]=1);var mi=xb(""),ni=!1,oi,pi=!1;pi=!0;
oi=pi;var qi,ri=!1;qi=ri;var si,ti=!1;si=ti;hi.Cd="www.googletagmanager.com";var ui=""+hi.Cd+(oi?"/gtag/js":"/gtm.js"),vi=null,wi=null,xi={},yi={};function zi(){var a=ii.sequence||1;ii.sequence=a+1;return a}hi.mk="true";var Ai="";hi.Sg=Ai;var Bi=new function(){this.j="";this.H=!1;this.D=0;this.R=this.vb=this.Z=this.Pa=this.M=""};function Ci(){var a=Bi.M.length;return Bi.M[a-1]==="/"?Bi.M.substring(0,a-1):Bi.M}function Di(){return Bi.H&&Bi.D!==1}function Ei(a){for(var b={},c=na(a.split("|")),d=c.next();!d.done;d=c.next())b[d.value]=!0;return b}var Fi=new ub,Gi={},Hi={},Ki={name:hi.hb,set:function(a,b){h(Jb(a,b),Gi);Ii()},get:function(a){return Ji(a,2)},reset:function(){Fi=new ub;Gi={};Ii()}};function Ji(a,b){return b!=2?Fi.get(a):Li(a)}function Li(a,b){var c=a.split(".");b=b||[];for(var d=Gi,e=0;e<c.length;e++){if(d===null)return!1;if(d===void 0)break;d=d[c[e]];if(b.indexOf(d)!==-1)return}return d}function Mi(a,b){Hi.hasOwnProperty(a)||(Fi.set(a,b),h(Jb(a,b),Gi),Ii())}
function Ni(){for(var a=["gtm.allowlist","gtm.blocklist","gtm.whitelist","gtm.blacklist","tagTypeBlacklist"],b=0;b<a.length;b++){var c=a[b],d=Ji(c,1);if(Array.isArray(d)||Ta(d))d=h(d);Hi[c]=d}}function Ii(a){z(Hi,function(b,c){Fi.set(b,c);h(Jb(b),Gi);h(Jb(b,c),Gi);a&&delete Hi[b]})}function Oi(a,b){var c,d=(b===void 0?2:b)!==1?Li(a):Fi.get(a);Qa(d)==="array"||Qa(d)==="object"?c=h(d):c=d;return c};
var Pi=function(a,b,c){if(!c)return!1;var d=c.selector_type,e=String(c.value),f;if(d==="js_variable"){e=e.replace(/\["?'?/g,".").replace(/"?'?\]/g,"");for(var g=e.split(","),k=0;k<g.length;k++){var m=g[k].trim();if(m){if(Gb(m,"dataLayer."))f=Ji(m.substring(10));else{var n=m.split(".");f=E[n.shift()];for(var p=0;p<n.length;p++)f=f&&f[n[p]]}if(f!==void 0)break}}}else if(d==="css_selector"&&lh)try{var q=kh(e);if(q&&q.length>0){f=[];for(var r=0;r<q.length&&r<(b==="email"||b==="phone_number"?5:1);r++)f.push(Fc(q[r])||
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -789,7 +789,7 @@
Q(52);
Q(23);Q(42);
Q(80);Q(93);Q(94);Q(91);Q(82);Q(49);Q(51);Q(61);Q(13);Q(48);
Q(68);function T(a){return!!fi[a]}var hi={},ii=E.google_tag_manager=E.google_tag_manager||{};hi.Rg="48r0";hi.oe=Number("0")||0;hi.hb="dataLayer";hi.fn="ChAI8LjAtgYQ8/fl7L7C6eFxEiUA4dkr/W2Ci5GiBjHO0vdJUTE3xMNnsrC+S131158/EjAOtQ7PGgILyg\x3d\x3d";var ji={__cl:1,__ecl:1,__ehl:1,__evl:1,__fal:1,__fil:1,__fsl:1,__hl:1,__jel:1,__lcl:1,__sdl:1,__tl:1,__ytl:1},ki={__paused:1,__tg:1},li;for(li in ji)ji.hasOwnProperty(li)&&(ki[li]=1);var mi=xb(""),ni=!1,oi,pi=!1;pi=!0;
Q(68);function T(a){return!!fi[a]}var hi={},ii=E.google_tag_manager=E.google_tag_manager||{};hi.Rg="48r0";hi.oe=Number("0")||0;hi.hb="dataLayer";hi.fn="ChAI8LjAtgYQ8/fl7L7C6eFxEiUA4dkr/cvOhNsViDCALixWZ/sVPNTj/b8fN2nv84v1MvW7UqGoGgI8Ow\x3d\x3d";var ji={__cl:1,__ecl:1,__ehl:1,__evl:1,__fal:1,__fil:1,__fsl:1,__hl:1,__jel:1,__lcl:1,__sdl:1,__tl:1,__ytl:1},ki={__paused:1,__tg:1},li;for(li in ji)ji.hasOwnProperty(li)&&(ki[li]=1);var mi=xb(""),ni=!1,oi,pi=!1;pi=!0;
oi=pi;var qi,ri=!1;qi=ri;var si,ti=!1;si=ti;hi.Cd="www.googletagmanager.com";var ui=""+hi.Cd+(oi?"/gtag/js":"/gtm.js"),vi=null,wi=null,xi={},yi={};function zi(){var a=ii.sequence||1;ii.sequence=a+1;return a}hi.mk="true";var Ai="";hi.Sg=Ai;var Bi=new function(){this.j="";this.H=!1;this.D=0;this.R=this.vb=this.Z=this.Pa=this.M=""};function Ci(){var a=Bi.M.length;return Bi.M[a-1]==="/"?Bi.M.substring(0,a-1):Bi.M}function Di(){return Bi.H&&Bi.D!==1}function Ei(a){for(var b={},c=na(a.split("|")),d=c.next();!d.done;d=c.next())b[d.value]=!0;return b}var Fi=new ub,Gi={},Hi={},Ki={name:hi.hb,set:function(a,b){h(Jb(a,b),Gi);Ii()},get:function(a){return Ji(a,2)},reset:function(){Fi=new ub;Gi={};Ii()}};function Ji(a,b){return b!=2?Fi.get(a):Li(a)}function Li(a,b){var c=a.split(".");b=b||[];for(var d=Gi,e=0;e<c.length;e++){if(d===null)return!1;if(d===void 0)break;d=d[c[e]];if(b.indexOf(d)!==-1)return}return d}function Mi(a,b){Hi.hasOwnProperty(a)||(Fi.set(a,b),h(Jb(a,b),Gi),Ii())}
function Ni(){for(var a=["gtm.allowlist","gtm.blocklist","gtm.whitelist","gtm.blacklist","tagTypeBlacklist"],b=0;b<a.length;b++){var c=a[b],d=Ji(c,1);if(Array.isArray(d)||Ta(d))d=h(d);Hi[c]=d}}function Ii(a){z(Hi,function(b,c){Fi.set(b,c);h(Jb(b),Gi);h(Jb(b,c),Gi);a&&delete Hi[b]})}function Oi(a,b){var c,d=(b===void 0?2:b)!==1?Li(a):Fi.get(a);Qa(d)==="array"||Qa(d)==="object"?c=h(d):c=d;return c};
var Pi=function(a,b,c){if(!c)return!1;var d=c.selector_type,e=String(c.value),f;if(d==="js_variable"){e=e.replace(/\["?'?/g,".").replace(/"?'?\]/g,"");for(var g=e.split(","),k=0;k<g.length;k++){var m=g[k].trim();if(m){if(Gb(m,"dataLayer."))f=Ji(m.substring(10));else{var n=m.split(".");f=E[n.shift()];for(var p=0;p<n.length;p++)f=f&&f[n[p]]}if(f!==void 0)break}}}else if(d==="css_selector"&&lh)try{var q=kh(e);if(q&&q.length>0){f=[];for(var r=0;r<q.length&&r<(b==="email"||b==="phone_number"?5:1);r++)f.push(Fc(q[r])||
Expand Down
Loading

0 comments on commit f22362a

Please sign in to comment.