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

    Neuronpedia

    INDEX
    Explanations

    always or often

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    possibly
    0.54
     किंतु
    0.46
    했고
    0.45
    겠지만
    0.45
    likely
    0.45
    роятно
    0.45
    indicated
    0.43
     Possibly
    0.43
    skip
    0.43
     тобто
    0.43
    POSITIVE LOGITS
     stets
    0.59
     wielu
    0.53
     zawsze
    0.53
     항상
    0.52
     mangas
    0.52
     často
    0.51
     souvent
    0.51
     altijd
    0.50
     spesso
    0.49
     sempre
    0.48
    Act Density 0.001%

    No Known Activations