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

    Neuronpedia

    INDEX
    Explanations

    phrases and words emphasizing self-reference and identity in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iastes
    -0.77
     uſed
    -0.65
     Monfieur
    -0.63
    بوابة
    -0.61
     lätt
    -0.60
     whoſe
    -0.60
     CURIAM
    -0.60
     sauvages
    -0.59
     hunne
    -0.59
     Chriftian
    -0.59
    POSITIVE LOGITS
     itself
    1.47
    itself
    1.35
     Itself
    1.29
    本身
    0.94
     самого
    0.88
     herself
    0.87
     himself
    0.87
     themselves
    0.85
    themselves
    0.84
    Сам
    0.81
    Act Density 0.197%

    No Known Activations