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

    Neuronpedia

    INDEX
    Explanations

    occurrences of the German word "zu" and its variations, indicating a focus on infinitive constructions or expressions involving purpose

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Majefty
    -0.89
     raiſ
    -0.87
     itſelf
    -0.86
     poffible
    -0.86
     Efq
    -0.84
     ſeveral
    -0.84
     Drapeau
    -0.81
    PyExc
    -0.80
     uſed
    -0.80
     AssemblyCulture
    -0.79
    POSITIVE LOGITS
     zu
    3.10
     Zu
    1.82
    Zu
    1.78
    zu
    1.44
     zum
    1.43
     zur
    1.34
     ZU
    1.20
    ZU
    0.98
    Zum
    0.95
     anzu
    0.94
    Act Density 0.014%

    No Known Activations