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

    Neuronpedia

    INDEX
    Explanations

    specific diacritical marks and special characters in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     hot
    -0.62
     ban
    -0.56
    Revenir
    -0.56
     
    -0.55
     anti
    -0.53
    zuführen
    -0.53
    styleType
    -0.53
     del
    -0.53
     data
    -0.52
     pan
    -0.52
    POSITIVE LOGITS
     itſelf
    1.02
     Houſe
    0.83
    thâu
    0.83
     himſelf
    0.81
     themſelves
    0.79
     neceſſ
    0.79
    InjectAttribute
    0.79
     myſelf
    0.78
     feroit
    0.77
     ainfi
    0.77
    Act Density 0.451%

    No Known Activations