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

    Neuronpedia

    INDEX
    Explanations

    type

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ร
    2.20
    एल
    1.95
     だけ
    1.89
    up
    1.86
    आर
    1.74
     berbentuk
    1.74
    ificantly
    1.72
     있으면
    1.72
    plik
    1.71
    ương
    1.70
    POSITIVE LOGITS
     kelamin
    2.16
    ת
    2.13
    ا
    2.11
    ভেদ
    1.80
    eled
    1.78
    ı
    1.75
     рода
    1.74
     accolades
    1.73
     grained
    1.73
    า
    1.65
    Act Density 0.382%

    No Known Activations