© 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. 7906
    Prev
    Next
    INDEX
    Explanations

    This neuron isn’t picking out any particular word or concept at all but is acting like a built-in “token counter” – its activation simply falls off steadily as you move through the text, so it effectively marks where you are in the document.

    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
    糟糕
    0.77
    ı
    0.73
    นี่
    0.73
     referred
    0.69
     hakkında
    0.68
     afar
    0.66
    ч
    0.66
    ل
    0.65
     averse
    0.65
    ব
    0.65
    POSITIVE LOGITS
    вершена
    0.80
    theoretic
    0.79
    moveTo
    0.78
     cubierta
    0.78
     Obstet
    0.77
    actose
    0.77
     socialization
    0.75
    antoin
    0.74
     ပြ
    0.73
    OLOGY
    0.72
    Activations Density 0.000%

    No Known Activations