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

    Neuronpedia

    INDEX
    Explanations

    default settings or values

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    odob
    0.40
    षिक
    0.38
     सच्चा
    0.37
    uric
    0.37
    ၤ
    0.37
    κτήθηκε
    0.36
    ฬ
    0.36
    തന്ത്ര
    0.36
     swollen
    0.35
    圻
    0.35
    POSITIVE LOGITS
     default
    4.31
    default
    3.92
     Default
    3.81
    默认
    3.67
    Default
    3.61
     defaults
    3.45
    デフォルト
    3.45
     DEFAULT
    3.11
     defaulted
    3.11
     Defaults
    3.02
    Act Density 0.243%

    No Known Activations