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

    Neuronpedia

    INDEX
    Explanations

    vital role

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    扮演
    0.71
     Roles
    0.68
    roles
    0.67
    Roles
    0.66
     roles
    0.63
     نقش
    0.63
     роли
    0.57
     peran
    0.57
     ruolo
    0.52
     역할
    0.49
    POSITIVE LOGITS
    OLE
    0.59
    ole
    0.57
     rule
    0.55
    ô
    0.54
    Ô
    0.52
     ole
    0.52
    rule
    0.49
    Ole
    0.48
     ро
    0.47
    Ро
    0.47
    Act Density 0.007%

    No Known Activations