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

    Neuronpedia

    INDEX
    Explanations

    parentheses or numbers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Carry
    0.61
     Morn
    0.59
    funeral
    0.57
     Morning
    0.57
     halten
    0.56
     Staying
    0.56
     Carry
    0.55
    aines
    0.55
    onha
    0.55
    Morning
    0.54
    POSITIVE LOGITS
    IAB
    0.65
    qm
    0.60
    QF
    0.57
    탑
    0.57
     goût
    0.56
    ях
    0.55
     Qui
    0.55
     pong
    0.54
    кт
    0.54
     Pab
    0.54
    Act Density 0.193%

    No Known Activations