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

    Neuronpedia

    INDEX
    Explanations

    like this:

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Impossible
    0.51
    Guarantee
    0.51
    磡
    0.49
     Конгрегация
    0.47
     ضمان
    0.47
    🚨
    0.46
     Еўро
    0.46
    CheckingType
    0.46
    絶対に
    0.46
    ರಿ
    0.45
    POSITIVE LOGITS
     three
    0.46
     his
    0.44
     despite
    0.43
     seven
    0.42
     yr
    0.41
     six
    0.40
     somewhat
    0.40
     several
    0.40
     Sun
    0.40
     Despite
    0.40
    Act Density 0.010%

    No Known Activations