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

    Neuronpedia

    INDEX
    Explanations

    control, peaceful, need

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     {
    0.57
    site
    0.54
    setup
    0.54
    smaller
    0.50
    service
    0.49
    விட்ட
    0.49
    _{
    0.48
    module
    0.48
    and
    0.48
    circuit
    0.47
    POSITIVE LOGITS
     guilds
    0.50
     енер
    0.49
     accett
    0.46
     अव
    0.46
     multiplications
    0.46
     crad
    0.46
     cytot
    0.45
     музи
    0.45
     άν
    0.44
     Пі
    0.44
    Act Density 0.001%

    No Known Activations