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

    Neuronpedia

    INDEX
    Explanations

    brightness, fainter, changes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ية
    0.74
    ة
    0.67
    дцать
    0.61
    ab
    0.58
     r
    0.55
    ing
    0.54
    carpeta
    0.54
     ر
    0.53
    ával
    0.53
    urro
    0.52
    POSITIVE LOGITS
     consortium
    0.58
     flagship
    0.57
    .
    0.55
     telev
    0.55
     peti
    0.55
    ل
    0.55
     способны
    0.54
     participan
    0.54
    ล
    0.54
     relay
    0.53
    Act Density 0.001%

    No Known Activations