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

    Neuronpedia

    INDEX
    Explanations

    behavior of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    蕺
    -2.13
    跸
    -1.95
     Anſ
    -1.95
    砗
    -1.91
    能让
    -1.89
    镡
    -1.85
     奶油
    -1.85
     toalla
    -1.82
    鮡
    -1.80
    ین
    -1.76
    POSITIVE LOGITS
     of
    2.03
    {
    1.94
     With
    1.93
     Also
    1.87
     Certainly
    1.84
     These
    1.80
     Should
    1.80
    on
    1.77
    We
    1.74
    b
    1.73
    Act Density 0.038%

    No Known Activations