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

    Neuronpedia

    INDEX
    Explanations

    before this, anymore

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     você
    0.58
    咱们
    0.58
     you
    0.56
    you
    0.54
     &
    0.53
     your
    0.51
    你
    0.50
    我们要
    0.50
     teie
    0.49
     we
    0.49
    POSITIVE LOGITS
     hadn
    0.69
     habían
    0.68
     había
    0.63
     unspoken
    0.57
     Lately
    0.55
     seemed
    0.54
     জানিয়েছিলেন
    0.54
     humiliation
    0.50
     somehow
    0.48
     কিংবা
    0.48
    Act Density 0.058%

    No Known Activations