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

    Neuronpedia

    INDEX
    Explanations

    explore investigating the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Всего
    0.40
     fuoco
    0.39
     सबकी
    0.38
    ிழமை
    0.37
     သူ့
    0.37
     nws
    0.37
     membuatnya
    0.36
     unusable
    0.36
     이름을
    0.36
     seus
    0.36
    POSITIVE LOGITS
     how
    0.86
    how
    0.80
     bagaimana
    0.80
     hvordan
    0.75
    如何
    0.66
     cómo
    0.66
     mengapa
    0.64
     কিভাবে
    0.63
    The
    0.62
     why
    0.61
    Act Density 0.011%

    No Known Activations