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

    Neuronpedia

    INDEX
    Explanations

    lived, Grand, Master, father, Lyra

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     aren
    0.88
     особливо
    0.76
     especially
    0.73
     put
    0.70
     begr
    0.70
    明け
    0.69
     argue
    0.69
     કબ
    0.68
     particularly
    0.68
    狡
    0.68
    POSITIVE LOGITS
     destes
    0.88
    Man
    0.84
    vič
    0.83
    ármaz
    0.82
    Ł
    0.81
    acup
    0.81
    кла
    0.79
     tejto
    0.79
     această
    0.78
    {
    0.78
    Act Density 0.002%

    No Known Activations