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

    Neuronpedia

    INDEX
    Explanations

    new words do not exist yet

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     advers
    0.80
    t
    0.79
    m
    0.78
     Closure
    0.77
    轳
    0.73
    rified
    0.73
    w
    0.73
     geometria
    0.71
     armada
    0.70
    í
    0.70
    POSITIVE LOGITS
    בית
    0.74
    ")}}
    0.70
    ות
    0.70
     måle
    0.68
    کي
    0.68
    elmi
    0.67
     scuole
    0.66
    Dat
    0.66
     शिक्षण
    0.66
     گی۔
    0.66
    Act Density 0.001%

    No Known Activations