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

    Neuronpedia

    INDEX
    Explanations

    external validation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    で
    0.95
    {
    0.90
    があり
    0.90
    piration
    0.84
    起源
    0.84
    íte
    0.81
     evasion
    0.78
    czeniu
    0.78
    以來
    0.77
    ication
    0.76
    POSITIVE LOGITS
    ंत्रिकी
    1.13
    ManagerProxy
    1.09
    Alumno
    1.05
    ಧಿಕ
    1.03
     complej
    1.02
    ی
    1.00
    lı
    0.99
     mango
    0.99
     כך
    0.97
    ಾರ
    0.96
    Act Density 0.002%

    No Known Activations