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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
     not
    -1.27
     at
    -1.18
     there
    -1.17
     then
    -1.09
     where
    -1.08
     detener
    -1.08
     which
    -1.07
     whose
    -1.05
     this
    -1.05
     if
    -1.05
    POSITIVE LOGITS
     demokra
    1.09
     kader
    1.01
    やはり
    1.01
    缡
    1.00
    fähigkeit
    0.99
    ⽼
    0.98
     間取り
    0.98
     numeros
    0.97
    ление
    0.97
     Frameworks
    0.97
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.