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

    Neuronpedia

    INDEX
    Explanations

    predicting outcomes or conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    可是
    0.44
    hormat
    0.41
     ভালো
    0.41
    ської
    0.40
    ومه
    0.40
     gimana
    0.40
     magari
    0.40
     showcased
    0.38
    之类的
    0.38
     whatnot
    0.38
    POSITIVE LOGITS
    எ
    0.43
     predicting
    0.40
     ensues
    0.39
     relapse
    0.39
     _:
    0.38
     removes
    0.38
     enzimas
    0.38
     nincs
    0.38
     successor
    0.38
    使用
    0.37
    Act Density 0.003%

    No Known Activations