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

    Neuronpedia

    INDEX
    Explanations

    terms indicating ease of use or simplicity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Inscrivez
    -0.70
    GenerationType
    -0.56
     cérami
    -0.56
     especiais
    -0.56
     spéciaux
    -0.55
    irsch
    -0.55
     feux
    -0.54
     engraçadas
    -0.53
     Substanz
    -0.53
     sauvages
    -0.53
    POSITIVE LOGITS
     easily
    4.26
    easily
    4.07
     Easily
    3.69
    Easily
    3.44
     fácilmente
    3.20
     facilmente
    3.04
     facilement
    2.92
     readily
    2.59
     łatwo
    1.90
     легко
    1.88
    Act Density 0.075%

    No Known Activations