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

    Neuronpedia

    INDEX
    Explanations

    created by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ocirc
    0.70
     probability
    0.67
     (
    0.63
     negó
    0.62
     namespace
    0.61
     documents
    0.60
     destiny
    0.59
    ,
    0.59
    probability
    0.57
    ogni
    0.57
    POSITIVE LOGITS
    сибо
    0.86
    οντας
    0.83
    ভেদ
    0.81
     제공
    0.80
    erreur
    0.80
     excellently
    0.79
    urando
    0.79
     entièrement
    0.79
     utiliser
    0.79
    ঞ্জাম
    0.78
    Act Density 0.136%

    No Known Activations