Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    several

    np_max-act · gemini-2.0-flash

    The neuron specifically activates on numerical tokens—particularly decimal numbers or measurements.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     disqualified
    -0.07
    Accent
    -0.07
    (mappedBy
    -0.07
    hazi
    -0.06
    /fire
    -0.06
    _bel
    -0.06
    _more
    -0.06
    QualifiedName
    -0.06
    .annotations
    -0.06
     modelName
    -0.06
    POSITIVE LOGITS
     poté
    0.06
    uencia
    0.06
    Ģ
    0.06
     çevre
    0.06
     temptation
    0.06
     Lair
    0.06
     taille
    0.06
     screws
    0.06
    इ
    0.05
    environment
    0.05
    Act Density 0.037%

    No Known Activations