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

    Neuronpedia

    INDEX
    Explanations

    dates and times

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    et
    0.53
    ar
    0.41
    o
    0.41
    an
    0.40
    e
    0.38
    es
    0.36
    en
    0.36
    us
    0.35
    a
    0.35
    er
    0.34
    POSITIVE LOGITS
     is
    0.32
    O
    0.30
    ﺃ
    0.30
    했습니다
    0.29
    Ч
    0.28
    encias
    0.28
     worse
    0.27
     പോലീസ്
    0.27
     يوم
    0.27
     discurso
    0.26
    Act Density 0.004%

    No Known Activations