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

    Neuronpedia

    INDEX
    Explanations

    physical descriptors

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     handed
    1.88
     headed
    1.79
     sided
    1.73
     minded
    1.71
     eyed
    1.66
    headed
    1.62
    handed
    1.61
     faced
    1.51
    sided
    1.50
    minded
    1.45
    POSITIVE LOGITS
     каттоо
    0.38
    unless
    0.38
    orsum
    0.38
    ELY
    0.37
    黨
    0.35
    ajno
    0.35
    ishma
    0.35
    жу
    0.35
     zert
    0.35
    ManagerState
    0.34
    Act Density 0.011%

    No Known Activations