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

    Neuronpedia

    INDEX
    Explanations

    activated or not

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     and
    -1.25
     estrenar
    -0.92
    将来
    -0.90
    And
    -0.88
     ٢
    -0.83
    lijkheid
    -0.82
     другие
    -0.82
    こういう
    -0.80
     シル
    -0.77
     loginUser
    -0.76
    POSITIVE LOGITS
     actual
    1.77
     yet
    1.65
     but
    1.60
     אלא
    1.44
     actually
    1.41
     nor
    1.41
    實際
    1.34
     itself
    1.30
    实际
    1.26
    而是
    1.23
    Act Density 0.082%

    No Known Activations