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

    Neuronpedia

    INDEX
    Explanations

    again

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :
    0.44
    ay
    0.42
    é
    0.42
    ol
    0.39
    ur
    0.38
    as
    0.37
    ie
    0.37
    ໋
    0.37
    ra
    0.36
    式
    0.36
    POSITIVE LOGITS
     again
    0.56
     novamente
    0.54
     nuevamente
    0.54
     opět
    0.51
     nuovamente
    0.49
     yine
    0.49
     опять
    0.47
     पुन्हा
    0.46
     Again
    0.45
     أيضا
    0.45
    Act Density 0.440%

    No Known Activations