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

    Neuronpedia

    INDEX
    Explanations

    the word "ein" and its variations, focusing on different forms of "one" or "a" in German

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     StatelessWidget
    -0.69
     "..\..\..\
    -0.69
     myſelf
    -0.68
     itſelf
    -0.68
     Oss
    -0.67
    Personendaten
    -0.66
     ―――――
    -0.65
    ARGS
    -0.65
     iſt
    -0.65
     Leto
    -0.65
    POSITIVE LOGITS
     ein
    1.90
     Ein
    1.46
    Ein
    1.40
     einen
    1.31
     een
    1.27
     eine
    1.13
    ein
    1.04
    Een
    0.99
     einem
    0.99
     Een
    0.98
    Act Density 0.018%

    No Known Activations