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

    Neuronpedia

    INDEX
    Explanations

    introducing what follows

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     कितने
    0.70
    そもそも
    0.65
    ljenje
    0.64
     উঠেছিল
    0.60
     Kalau
    0.59
     öss
    0.59
    δρο
    0.57
     alte
    0.56
     tört
    0.56
     kõr
    0.56
    POSITIVE LOGITS
     below
    3.94
    below
    3.36
    Below
    3.25
     Below
    3.23
    以下
    2.99
     abaixo
    2.99
    下面
    2.92
    下面的
    2.89
     BELOW
    2.85
     ниже
    2.82
    Act Density 0.871%

    No Known Activations