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

    Neuronpedia

    INDEX
    Explanations

    someone

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    そして
    0.45
     however
    0.44
     azonban
    0.43
    とその
    0.43
     également
    0.43
     ancak
    0.43
     tačiau
    0.42
     Однако
    0.42
     HOWEVER
    0.40
     però
    0.39
    POSITIVE LOGITS
    .,
    0.67
    someone
    0.53
     ktoś
    0.52
     someone
    0.50
    Если
    0.49
    有人
    0.48
     iemand
    0.48
     alguém
    0.47
     เงี้ย
    0.46
    Someone
    0.46
    Act Density 0.030%

    No Known Activations