Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment

Vinay Setty

Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment

Correct Metadata for

Use this form to create a GitHub issue with structured data describing the correction. You will need a GitHub account. Once you create that issue, the correction will be reviewed by a staff member.

⚠️ Mobile Users: Submitting this form to create a new issue will only work with github.com, not the GitHub Mobile app.

Important: The Anthology treat PDFs as authoritative. Please use this form only to correct data that is out of line with the PDF. See our corrections guidelines if you need to change the PDF.

Title Adjust the title. Retain tags such as <fixed-case>.

Authors Adjust author names and order to match the PDF.

Abstract Correct abstract if needed. Retain XML formatting tags such as <tex-math>. You may use ... for bold, ... for italic, ... for underline, <sc>...</sc> for small-caps, <tt>...<tt> for typewriter text, <url>...</url> for URLs, <a href=...> for hyperlinks, and <par/> for paragraph breaks.

Verification against PDF Ensure that the new title/authors match the snapshot below. (If there is no snapshot or it is too small, consult the PDF.)

Authors concatenated from the text boxes above:

ALL author names match the snapshot above—including middle initials, hyphens, and accents.

Abstract

We reproduce the canonical GPT-2-small “greater-than” circuit in its original stack and independently, then test how far it generalizes across eight models (124M–12B) and to numerical entailment. The reported numbers replicate, and the late-MLP structure recurs at matched relative depth well above a size-matched null. Sparse faithfulness is high everywhere too, but only when the edge budget scales with the model’s own graph; since these graphs differ 96-fold in size, a fixed budget makes faithfulness appear to collapse with scale. On numerical entailment, no circuit-bearing model passes the 0.6 gate, while the one model that solves the task well cannot be posed the circuit’s own task.

Anthology ID:: 2026.blackboxnlp-1.7
Volume:: Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP
Month:: October
Year:: 2026
Address:: Budapest, Hungary
Editors:: Dana Arad, Yonatan Belinkov, Hanjie Chen, Nils Feldhus, Aaron Mueller, Hosein Mohebbi, Gabriele Sarti, Martin Tutek
Venues:: BlackboxNLP | WS
SIG:
Publisher:: Association for Computational Linguistics
Note:
Pages:: 90–102
Language:
URL:: https://aclanthology.org/2026.blackboxnlp-1.7/
DOI:
Bibkey:
Cite (ACL):: Vinay Setty. 2026. Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment. In Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 90–102, Budapest, Hungary. Association for Computational Linguistics.
Cite (Informal):: Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment (Setty, BlackboxNLP 2026)
Copy Citation:
PDF:: https://aclanthology.org/2026.blackboxnlp-1.7.pdf

PDF Cite Search Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@inproceedings{setty-2026-greater,
    title = "Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment",
    author = "Setty, Vinay",
    editor = "Arad, Dana  and
      Belinkov, Yonatan  and
      Chen, Hanjie  and
      Feldhus, Nils  and
      Mueller, Aaron  and
      Mohebbi, Hosein  and
      Sarti, Gabriele  and
      Tutek, Martin",
    booktitle = "Proceedings of the 9th {B}lackbox{NLP} Workshop: Analyzing and Interpreting Neural Networks for {NLP}",
    month = oct,
    year = "2026",
    address = "Budapest, Hungary",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.blackboxnlp-1.7/",
    pages = "90--102",
    ISBN = "979-8-89176-476-7",
    abstract = "We reproduce the canonical GPT-2-small ``greater-than'' circuit in its original stack and independently, then test how far it generalizes across eight models (124M{--}12B) and to numerical entailment. The reported numbers replicate, and the late-MLP structure recurs at matched relative depth well above a size-matched null. Sparse faithfulness is high everywhere too, but only when the edge budget scales with the model{'}s own graph; since these graphs differ 96-fold in size, a fixed budget makes faithfulness appear to collapse with scale. On numerical entailment, no circuit-bearing model {passes the 0.6 gate}, while the one model that solves the task well cannot be posed the circuit{'}s own task."
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="setty-2026-greater">
    <titleInfo>
        <title>Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Vinay</namePart>
        <namePart type="family">Setty</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Dana</namePart>
            <namePart type="family">Arad</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yonatan</namePart>
            <namePart type="family">Belinkov</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hanjie</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nils</namePart>
            <namePart type="family">Feldhus</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Aaron</namePart>
            <namePart type="family">Mueller</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hosein</namePart>
            <namePart type="family">Mohebbi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Gabriele</namePart>
            <namePart type="family">Sarti</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Martin</namePart>
            <namePart type="family">Tutek</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-476-7</identifier>
    </relatedItem>
    <abstract>We reproduce the canonical GPT-2-small “greater-than” circuit in its original stack and independently, then test how far it generalizes across eight models (124M–12B) and to numerical entailment. The reported numbers replicate, and the late-MLP structure recurs at matched relative depth well above a size-matched null. Sparse faithfulness is high everywhere too, but only when the edge budget scales with the model’s own graph; since these graphs differ 96-fold in size, a fixed budget makes faithfulness appear to collapse with scale. On numerical entailment, no circuit-bearing model passes the 0.6 gate, while the one model that solves the task well cannot be posed the circuit’s own task.</abstract>
    <identifier type="citekey">setty-2026-greater</identifier>
    <location>
        <url>https://aclanthology.org/2026.blackboxnlp-1.7/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>90</start>
            <end>102</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Conference Proceedings
%T Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment
%A Setty, Vinay
%Y Arad, Dana
%Y Belinkov, Yonatan
%Y Chen, Hanjie
%Y Feldhus, Nils
%Y Mueller, Aaron
%Y Mohebbi, Hosein
%Y Sarti, Gabriele
%Y Tutek, Martin
%S Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-476-7
%F setty-2026-greater
%X We reproduce the canonical GPT-2-small “greater-than” circuit in its original stack and independently, then test how far it generalizes across eight models (124M–12B) and to numerical entailment. The reported numbers replicate, and the late-MLP structure recurs at matched relative depth well above a size-matched null. Sparse faithfulness is high everywhere too, but only when the edge budget scales with the model’s own graph; since these graphs differ 96-fold in size, a fixed budget makes faithfulness appear to collapse with scale. On numerical entailment, no circuit-bearing model passes the 0.6 gate, while the one model that solves the task well cannot be posed the circuit’s own task.
%U https://aclanthology.org/2026.blackboxnlp-1.7/
%P 90-102

Download as File

Markdown (Informal)

[Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment](https://aclanthology.org/2026.blackboxnlp-1.7/) (Setty, BlackboxNLP 2026)

Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment (Setty, BlackboxNLP 2026)

ACL

Vinay Setty. 2026. Does the Greater-Than Circuit Generalize? A Reproducibility Study Across Models and Numerical Entailment. In Proceedings of the 9th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 90–102, Budapest, Hungary. Association for Computational Linguistics.