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

    Neuronpedia

    INDEX
    Explanations

    time duration

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     formally
    0.95
     форма
    0.91
     formal
    0.82
     Formal
    0.77
    Formal
    0.77
    form
    0.76
    formal
    0.75
     form
    0.71
     Form
    0.70
    Form
    0.67
    POSITIVE LOGITS
     before
    0.52
    before
    0.50
     sebelum
    0.50
     trước
    0.47
     antes
    0.46
     prije
    0.46
     قبل
    0.45
     voordat
    0.44
     innan
    0.43
     ennen
    0.42
    Act Density 0.001%

    No Known Activations