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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6245
    Prev
    Next
    INDEX
    Explanations

    page numbers, citations, technical identifiers

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires strongly on the numeric parts of academic citations (volume numbers, page ranges, and similar digits in references).

    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_31_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
    ávez
    0.57
    随着
    0.56
     نفسی
    0.55
     Ambiental
    0.54
     ಕನ್ನಡ
    0.51
    clickHandler
    0.51
    𝐺
    0.51
    elola
    0.50
     Isi
    0.49
    Direct
    0.49
    POSITIVE LOGITS
     page
    0.99
     pages
    0.95
    页
    0.95
     pg
    0.93
     xvii
    0.91
     Pages
    0.89
     صفحات
    0.89
     pagine
    0.89
    page
    0.88
     Page
    0.88
    Activations Density 0.187%

    No Known Activations