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

    Neuronpedia

    INDEX
    Explanations

    infinitely repeating patterns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :
    0.56
     are
    0.53
    ::
    0.53
    ;
    0.52
     and
    0.52
    '
    0.52
       
    0.51
    us
    0.51
    )
    0.51
          
    0.48
    POSITIVE LOGITS
     کہانی
    0.66
     करियर
    0.63
     showbiz
    0.61
     चर्चित
    0.59
     stardom
    0.58
     lackluster
    0.57
    Noeud
    0.57
     داستان
    0.55
     कहानी
    0.55
     lovable
    0.55
    Act Density 0.003%

    No Known Activations