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

    Neuronpedia

    INDEX
    Explanations

    list of instructions/notes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ,
    0.50
    :
    0.47
    nya
    0.47
     cylinders
    0.43
     detected
    0.42
     may
    0.42
    ."
    0.42
    导致
    0.41
    ;
    0.41
     responds
    0.41
    POSITIVE LOGITS
    我已经
    0.77
     আমি
    0.70
     제가
    0.68
     Итак
    0.65
    我
    0.65
    আমি
    0.64
     ನಾನು
    0.64
    我现在
    0.63
     আমার
    0.62
     मैंने
    0.62
    Act Density 0.002%

    No Known Activations