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

    Neuronpedia

    INDEX
    Explanations

    penetrate beyond

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Asking
    0.51
     Ask
    0.47
    Ask
    0.46
    Asking
    0.43
    asking
    0.43
     pergunt
    0.41
     спраши
    0.41
    Przy
    0.41
     Familiar
    0.40
    Ско
    0.40
    POSITIVE LOGITS
     anything
    0.51
     goose
    0.42
     whatever
    0.42
     gradient
    0.41
     stock
    0.39
     هزار
    0.38
     thousand
    0.37
    zent
    0.37
     ANYTHING
    0.36
     favor
    0.36
    Act Density 0.000%

    No Known Activations