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

    Neuronpedia

    INDEX
    Explanations

    past tense descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     of
    0.40
    0.36
    ule
    0.33
    أ
    0.33
    nent
    0.33
    ari
    0.32
    %\
    0.31
    ;
    0.30
    在
    0.30
     be
    0.30
    POSITIVE LOGITS
     doesn
    0.38
     dovrà
    0.37
    пи
    0.34
     olacaktır
    0.34
    場合は
    0.33
    ے
    0.33
     can
    0.33
    다는
    0.33
     وقتی
    0.33
    ро
    0.33
    Act Density 0.161%

    No Known Activations