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

    Neuronpedia

    INDEX
    Explanations

    markdown headings or lists

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     be
    0.95
     pre
    0.89
     open
    0.89
     element
    0.87
     d
    0.86
    ,
    0.85
    ',
    0.84
     own
    0.84
     end
    0.84
     simple
    0.83
    POSITIVE LOGITS
    Despite
    1.45
    According
    1.44
    Although
    1.41
    However
    1.37
     Также
    1.37
     Additionally
    1.34
    Asimismo
    1.33
    But
    1.32
     Además
    1.32
    While
    1.31
    Act Density 0.000%

    No Known Activations