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

    Neuronpedia

    INDEX
    Explanations

    breakdown and estimates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     død
    0.40
     rę
    0.39
    흐
    0.38
     eax
    0.38
    orrh
    0.37
     দৈর্ঘ
    0.37
     settim
    0.36
    始终
    0.36
    路线
    0.36
    Hence
    0.35
    POSITIVE LOGITS
     бы
    0.45
     their
    0.45
     characterizes
    0.41
     продукты
    0.40
     customized
    0.39
     их
    0.39
     Haber
    0.38
     tailor
    0.38
    ierta
    0.37
    utia
    0.37
    Act Density 0.000%

    No Known Activations