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

    Neuronpedia

    INDEX
    Explanations

    specific options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ，
    0.55
     ਅਤੇ
    0.53
    这个
    0.46
     અને
    0.44
    たくさん
    0.43
     всі
    0.42
    ान
    0.41
    ната
    0.41
     ，
    0.41
    ，“
    0.41
    POSITIVE LOGITS
     be
    0.64
    P
    0.59
    ق
    0.54
    _
    0.49
    2
    0.46
    ــــ
    0.45
     P
    0.43
    ير
    0.42
     kindred
    0.41
    U
    0.41
    Act Density 1.647%

    No Known Activations