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

    Neuronpedia

    INDEX
    Explanations

    Subject: lines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ায়
    0.80
    Precis
    0.75
    =
    0.72
    (\\
    0.72
    and
    0.72
    to
    0.71
    the
    0.70
    苄
    0.70
    provides
    0.69
    place
    0.69
    POSITIVE LOGITS
     thwarted
    1.01
     helfen
    0.97
     worsened
    0.91
     urg
    0.89
     stabbing
    0.88
     fraught
    0.88
     Optimized
    0.87
     doulou
    0.87
     boosting
    0.86
     ഒഴിവാ
    0.86
    Act Density 0.019%

    No Known Activations