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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a recurring action or event

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "whenever" indicating conditional or recurring situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    asse
    -0.80
    gur
    -0.71
    atory
    -0.70
    hash
    -0.69
    ettle
    -0.68
    atories
    -0.68
    nee
    -0.66
    him
    -0.65
    olves
    -0.65
    ruly
    -0.64
    POSITIVE LOGITS
    soever
    1.52
     else
    0.91
    wcsstore
    0.90
    entimes
    0.86
    ĸļ
    0.85
    £ı
    0.84
     thereafter
    0.84
    */(
    0.82
     oun
    0.81
    theless
    0.80
    Act Density 0.007%

    No Known Activations