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

    Neuronpedia

    INDEX
    Explanations

    specific items followed by a colon

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Р
    1.17
    ことから
    1.11
    ্শন
    1.11
    прият
    1.07
    Ма
    1.02
    ся
    1.02
    יות
    1.01
    ෙන
    0.98
    Не
    0.96
     Faites
    0.96
    POSITIVE LOGITS
     luxury
    1.03
     lur
    0.97
     institutional
    0.95
    €˜
    0.93
    irikan
    0.89
     jungle
    0.89
    ந
    0.89
     modest
    0.89
     frontline
    0.89
     phonon
    0.89
    Act Density 0.006%

    No Known Activations