© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 5573
    Prev
    Next
    INDEX
    Explanations

    This neuron is primarily responsive to numeric list markers and section-style numbers—that is, tokens containing digits used for numbered headings, list items, or other structured numeric labels.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     aort
    0.63
     it
    0.62
    𝖺
    0.62
     hbox
    0.61
    <strong>
    0.59
     cartera
    0.59
    <b>
    0.58
    apot
    0.58
    <i>
    0.57
     a
    0.57
    POSITIVE LOGITS
    </b>
    1.48
    </h2>
    1.44
    </strong>
    1.43
    </u>
    1.29
    </h4>
    1.22
    </code>
    1.19
    </h3>
    1.17
    </span>
    1.13
    </em>
    1.10
    ‬
    0.99
    Activations Density 0.261%

    No Known Activations