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

    Neuronpedia

    INDEX
    Explanations

    cardinal directions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    etrics
    -0.08
     town
    -0.08
    proto
    -0.07
     Darkness
    -0.07
     Accounting
    -0.07
     dro
    -0.07
     ign
    -0.07
    ỉnh
    -0.07
     Porter
    -0.06
    odule
    -0.06
    POSITIVE LOGITS
    ريل
    0.07
     mutation
    0.07
    ')==
    0.06
    -expanded
    0.06
    ,total
    0.06
     постав
    0.06
    ์โ
    0.06
     interes
    0.06
    ;


↵
    0.06
    [float
    0.06
    Act Density 0.013%

    No Known Activations