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

    Neuronpedia

    INDEX
    Explanations

    chosen word or concept

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     postice
    0.47
     polskich
    0.47
     российских
    0.46
     unor
    0.44
     wholesalers
    0.43
     whi
    0.43
     actively
    0.43
     международных
    0.42
     internationally
    0.42
     tzv
    0.42
    POSITIVE LOGITS
    Chosen
    0.50
     Chosen
    0.48
     выбран
    0.45
     столь
    0.44
    所
    0.44
     hapless
    0.43
     keseluruhan
    0.43
    кову
    0.42
     chosen
    0.41
    整个
    0.41
    Act Density 0.006%

    No Known Activations