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

    Neuronpedia

    INDEX
    Explanations

    character walk

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     shouldn
    0.98
    <unused99>
    0.96
     hasn
    0.92
    <unused89>
    0.91
     ብዙውን
    0.90
     needn
    0.89
     Está
    0.88
    }&=
    0.88
    <unused2157>
    0.87
     IMHO
    0.86
    POSITIVE LOGITS
     or
    0.67
    (
    0.65
     /
    0.63
    asse
    0.63
    টি
    0.62
    i
    0.61
     Ob
    0.59
     (*)
    0.58
     imaginable
    0.58
    more
    0.57
    Act Density 0.442%

    No Known Activations