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

    Neuronpedia

    INDEX
    Explanations

    phrases related to unusual, outlandish, or extreme situations

    oai_token-act-pair · gpt-3.5-turbo

    variations of the word "freak" and its derivatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    adr
    -0.87
    ournal
    -0.79
     conduc
    -0.78
    fman
    -0.75
    ccording
    -0.75
    ript
    -0.72
    eger
    -0.70
    idates
    -0.67
    İĭ
    -0.67
     subscript
    -0.66
    POSITIVE LOGITS
    ishly
    1.13
    onom
    0.92
    show
    0.82
    bum
    0.82
    ously
    0.81
    istically
    0.77
     freak
    0.72
    ety
    0.72
    iness
    0.71
    auld
    0.71
    Act Density 0.028%

    No Known Activations