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

    Neuronpedia

    INDEX
    Explanations

    news announcement titles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     čega
    2.14
    ➖
    1.93
    lı
    1.85
    úscula
    1.84
     oč
    1.83
     Ś
    1.79
    ম
    1.78
    л
    1.77
    mêmes
    1.73
     tego
    1.72
    POSITIVE LOGITS
    ি
    2.12
    i
    2.11
    least
    1.69
    though
    1.62
    রকম
    1.52
    Sexy
    1.47
    isActive
    1.47
     ease
    1.47
    Expl
    1.47
    ^{
    1.43
    Act Density 0.000%

    No Known Activations