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

    Neuronpedia

    INDEX
    Explanations

    Why bother

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     解説
    0.44
    容易
    0.42
     ಅಂಶ
    0.42
    Lik
    0.41
     inappropri
    0.41
     explan
    0.41
     tendencies
    0.41
     різні
    0.40
     strutt
    0.39
     unsuitable
    0.39
    POSITIVE LOGITS
     überhaupt
    0.77
     вообще
    0.69
     should
    0.66
     भला
    0.59
    わざ
    0.59
     bother
    0.57
     Should
    0.57
     needed
    0.57
     siquiera
    0.54
     اصلا
    0.54
    Act Density 0.008%

    No Known Activations