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

    Neuronpedia

    INDEX
    Explanations

    changing rapidly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     realizando
    0.61
     causando
    0.61
     doing
    0.57
     fazendo
    0.56
     haciendo
    0.54
     facendo
    0.53
     preparando
    0.53
     dando
    0.52
     manteniendo
    0.52
    Doing
    0.52
    POSITIVE LOGITS
     rapidly
    0.71
     quickly
    0.69
     slowly
    0.60
     rápidamente
    0.59
     rapidamente
    0.59
     steadily
    0.56
     swiftly
    0.54
     быстро
    0.52
     Quickly
    0.51
    quickly
    0.50
    Act Density 0.036%

    No Known Activations