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

    Neuronpedia

    INDEX
    Explanations

    from separation source

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    .
    -2.28
     of
    -1.99
    .'
    -1.90
     was
    -1.82
     "
    -1.79
     into
    -1.51
     pertenec
    -1.51
     a
    -1.50
    鬟
    -1.46
     放送
    -1.45
    POSITIVE LOGITS
     nôtre
    1.97
     galeri
    1.83
    𝐆
    1.76
    ᭝
    1.74
     araba
    1.73
    autres
    1.72
    hloromethane
    1.68
    媖
    1.67
     vilja
    1.67
    ﹍﹍
    1.63
    Act Density 0.033%

    No Known Activations