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

    Neuronpedia

    INDEX
    Explanations

    `_`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     from
    -1.95
     because
    -1.88
     so
    -1.81
     for
    -1.68
     and
    -1.67
     ขาว
    -1.63
     any
    -1.61
     Since
    -1.55
     after
    -1.52
     before
    -1.52
    POSITIVE LOGITS
    搋
    1.94
    螓
    1.86
    atów
    1.84
    肏
    1.79
    髏
    1.77
    1.62
    1.56
    袛
    1.55
     riks
    1.52
    ネート
    1.51
    Act Density 0.132%

    No Known Activations