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

    Neuronpedia

    INDEX
    Explanations

    continuation until a condition

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     разных
    0.80
     আজকাল
    0.79
    ثير
    0.77
     प्योर
    0.73
    了我
    0.72
     многих
    0.70
    的一些
    0.69
    有些
    0.69
    แต่ละ
    0.68
    やっぱり
    0.68
    POSITIVE LOGITS
     until
    5.55
    until
    5.13
     Until
    4.93
    Until
    4.75
     hasta
    4.06
    直到
    4.03
     till
    3.95
     até
    3.83
     jusqu
    3.73
     sampai
    3.40
    Act Density 0.122%

    No Known Activations