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

    Neuronpedia

    INDEX
    Explanations

    signed/unsigned variants

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    au
    0.48
     また
    0.42
    asak
    0.42
     freg
    0.42
     sii
    0.41
    lw
    0.41
    uig
    0.40
     tine
    0.40
     LWR
    0.40
    न्दु
    0.40
    POSITIVE LOGITS
    给她
    0.42
     Förderung
    0.40
    Saldo
    0.39
     Dtsch
    0.39
     môžete
    0.38
    Sov
    0.38
     Hydra
    0.38
    普通的
    0.38
     normalerweise
    0.38
    BALANCE
    0.38
    Act Density 0.005%

    No Known Activations