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

    Neuronpedia

    INDEX
    Explanations

    overwhelming, confusing, offensive

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    re
    0.54
    ны
    0.52
    مر
    0.51
    ien
    0.50
    ired
    0.50
    giveness
    0.50
    ik
    0.50
     t
    0.49
    йы
    0.49
    into
    0.49
    POSITIVE LOGITS
    -
    0.64
     dazz
    0.59
    고
    0.57
     Dla
    0.56
    ப்பூ
    0.55
     offens
    0.55
     startling
    0.53
     FIA
    0.52
     Quais
    0.51
     Cols
    0.51
    Act Density 0.095%

    No Known Activations