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

    Neuronpedia

    INDEX
    Explanations

    compared

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     خواہ
    0.68
     inescapable
    0.61
     walang
    0.61
     niemals
    0.60
     alltid
    0.59
     jamás
    0.58
    совой
    0.58
     pintor
    0.58
     perpetuated
    0.58
     predominate
    0.57
    POSITIVE LOGITS
     compared
    1.02
     Compared
    0.96
    compared
    0.90
    Compared
    0.86
    相比
    0.80
     having
    0.80
     considering
    0.80
     since
    0.79
     dibandingkan
    0.79
     after
    0.78
    Act Density 0.000%

    No Known Activations