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

    Neuronpedia

    INDEX
    Explanations

    concluding phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     başlayalım
    0.53
     ­
    0.51
    츨
    0.49
    ’:
    0.49
     …….
    0.49
    <strong>
    0.48
    텝
    0.48
     befindet
    0.48
    ，
    0.47
    ’।
    0.46
    POSITIVE LOGITS
     Hence
    1.03
     Therefore
    0.97
    Hence
    0.93
    Therefore
    0.90
     Saying
    0.80
     Ironically
    0.80
     Thus
    0.79
     所以
    0.78
     Certainly
    0.77
     Unless
    0.76
    Act Density 0.200%

    No Known Activations