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

    Neuronpedia

    INDEX
    Explanations

    variable assignments and list access

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ,$$
    0.46
    ,
    0.44
    vieve
    0.43
    fromi
    0.39
    ,【
    0.39
    perror
    0.39
    quele
    0.38
    ről
    0.38
    leştir
    0.38
    你了
    0.38
    POSITIVE LOGITS
     is
    0.85
     are
    0.80
    на
    0.71
     was
    0.68
     has
    0.63
     were
    0.61
    о
    0.59
     of
    0.57
     and
    0.57
     can
    0.56
    Act Density 2.178%

    No Known Activations