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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
    gradient
    -0.80
    loo
    -0.78
    onto
    -0.70
    zzo
    -0.68
    instead
    -0.67
    umo
    -0.64
     preference
    -0.64
    cest
    -0.64
    dayName
    -0.64
    oga
    -0.64
    POSITIVE LOGITS
     Wikileaks
    0.64
    ĸļ
    0.63
     Aviv
    0.62
    ENSE
    0.62
     Rica
    0.62
     DPRK
    0.62
     ver
    0.60
    Material
    0.60
     Nether
    0.59
     condemns
    0.59
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.