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

    Neuronpedia

    INDEX
    Explanations

    Single digits

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens (quantities and digits).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Seasons
    -0.08
    eco
    -0.07
    Confirm
    -0.06
     interception
    -0.06
    Ul
    -0.06
     seasoning
    -0.06
    ्रभ
    -0.06
     promotions
    -0.06
    breadcrumbs
    -0.06
     Dare
    -0.06
    POSITIVE LOGITS
    ۲۶
    0.06
     doivent
    0.06
    zed
    0.06
    -dd
    0.06
     whichever
    0.06
    hq
    0.06
     drm
    0.06
     jež
    0.06
     компанії
    0.06
    нила
    0.05
    Act Density 0.026%

    No Known Activations