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

    Neuronpedia

    INDEX
    Explanations

    retribution, Trump, pregnancy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     cuáles
    0.51
     сколько
    0.48
    Ї
    0.46
    Все
    0.45
    Ис
    0.45
     marvellous
    0.44
     लाहिड़ी
    0.44
     зовут
    0.43
    ilaian
    0.43
     多い
    0.43
    POSITIVE LOGITS
     with
    0.59
     for
    0.56
     of
    0.54
     on
    0.53
    type
    0.51
    l
    0.49
     einer
    0.48
     type
    0.47
     eine
    0.47
     from
    0.47
    Act Density 0.004%

    No Known Activations