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

    Neuronpedia

    INDEX
    Explanations

    specific code or url part

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fluff
    0.76
     fam
    0.71
     scrutin
    0.70
     myst
    0.68
    ப்பன்
    0.67
     premise
    0.67
     sanity
    0.67
     though
    0.66
     persona
    0.65
     frequent
    0.64
    POSITIVE LOGITS
    其余
    1.03
    上述
    0.99
     อื่น
    0.94
     其他
    0.93
     cGraph
    0.93
     tzw
    0.92
     żad
    0.89
    即
    0.88
     jedynie
    0.88
    .$\
    0.88
    Act Density 0.405%

    No Known Activations