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

    Neuronpedia

    INDEX
    Explanations

    private steps list diff hidden index values cumulative memo cum

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ﺪ
    0.68
     aérea
    0.64
    यॉर्क
    0.56
     alemão
    0.55
    ﺶ
    0.55
     piensan
    0.54
     alemán
    0.53
     mehrerer
    0.53
     perfeita
    0.52
     Надо
    0.52
    POSITIVE LOGITS
    il
    0.63
    ↵
    0.62
    ar
    0.57
    al
    0.55
    ant
    0.54
     (
    0.52
    el
    0.51
    et
    0.51
     de
    0.51
    en
    0.50
    Act Density 0.000%

    No Known Activations