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

    Neuronpedia

    INDEX
    Explanations

    Russian telegram terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     york
    0.46
     arithmetic
    0.44
     fill
    0.44
     five
    0.43
     analytical
    0.43
     black
    0.43
     čern
    0.42
     ascetic
    0.42
    漼
    0.41
    så
    0.41
    POSITIVE LOGITS
    тел
    0.48
    }^{
    0.47
    телей
    0.47
     União
    0.47
    }^{\
    0.46
     после
    0.46
    agues
    0.45
     TEL
    0.45
    тельностью
    0.44
    Tel
    0.44
    Act Density 0.007%

    No Known Activations