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

    Neuronpedia

    INDEX
    Explanations

    asking if you want more

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     invisible
    0.46
     mieux
    0.45
    变化的
    0.43
     empty
    0.43
     differently
    0.42
     accounting
    0.42
     uniqueness
    0.41
     someone
    0.41
     us
    0.40
     responsiveness
    0.40
    POSITIVE LOGITS
     destacado
    0.71
     Einige
    0.68
     seleccionado
    0.67
     хочу
    0.67
     quieren
    0.66
    want
    0.63
     हैज
    0.62
     einige
    0.61
    Want
    0.61
     quiser
    0.60
    Act Density 0.376%

    No Known Activations