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

    Neuronpedia

    INDEX
    Explanations

    freeCodeCamp, Scott's

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ます
    2.69
    ন
    2.16
    ون
    1.98
    ますが
    1.78
     marts
    1.75
    ين
    1.70
    みる
    1.67
     walnuts
    1.60
    ੁ
    1.58
    э
    1.57
    POSITIVE LOGITS
     пут
    1.62
    kan
    1.57
    es
    1.55
    aniti
    1.53
    ).
    1.42
     majeur
    1.42
    )$
    1.40
    verfahren
    1.40
    i
    1.39
    颯
    1.38
    Act Density 0.000%

    No Known Activations