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

    Neuronpedia

    INDEX
    Explanations

    absolutely not

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     బాగా
    0.60
     purely
    0.57
     주로
    0.56
     преимущественно
    0.55
    未必
    0.55
     полноцен
    0.55
     जास्त
    0.54
     நிறைய
    0.54
     mainly
    0.53
     terlalu
    0.53
    POSITIVE LOGITS
     ANYTHING
    1.09
     whatsoever
    1.01
    任何
    1.00
     EVER
    0.97
     anything
    0.91
     ANY
    0.91
     ever
    0.89
    Anything
    0.82
     anywhere
    0.80
     qualquer
    0.80
    Act Density 0.047%

    No Known Activations