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

    Neuronpedia

    INDEX
    Explanations

    Ask me about what firewall in wipe out

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ATT
    1.12
     both
    1.11
    ported
    1.01
     Ass
    1.01
     kass
    0.99
    hase
    0.99
     one
    0.98
     ot
    0.97
     oba
    0.96
     entrambi
    0.95
    POSITIVE LOGITS
    7
    1.97
    8
    1.93
    6
    1.93
    9
    1.82
    4
    1.69
    5
    1.68
    0
    1.49
    3
    1.46
    º
    1.27
    2
    1.26
    Act Density 0.576%

    No Known Activations