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

    Neuronpedia

    INDEX
    Explanations

    variable assignments in code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     for
    0.57
    ת
    0.52
    ቈ
    0.44
    ကျ
    0.44
    kad
    0.44
    ുകൊണ്ടാണ്
    0.43
    राबरी
    0.42
    社会
    0.42
    草
    0.42
    গত
    0.42
    POSITIVE LOGITS
    4
    0.70
    .
    0.58
    :
    0.55
    ,
    0.54
     все
    0.53
    ;
    0.52
     €,
    0.52
     例えば
    0.52
     muttered
    0.49
     vole
    0.49
    Act Density 0.070%

    No Known Activations