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

    Neuronpedia

    INDEX
    Explanations

    allow and permission

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ବା
    0.44
     ideales
    0.43
    ब
    0.42
    druk
    0.42
    तक
    0.41
     Узбеки
    0.41
    ب
    0.40
     eest
    0.39
    বায়
    0.39
    ুব
    0.38
    POSITIVE LOGITS
     Allow
    0.74
    Allow
    0.70
    ALLOW
    0.62
     allow
    0.59
    allow
    0.59
     Allowed
    0.57
     Allowing
    0.57
    允许
    0.56
     اجازه
    0.54
     erlaub
    0.51
    Act Density 0.005%

    No Known Activations