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

    Neuronpedia

    INDEX
    Explanations

    phrases referring to choices or options

    oai_token-act-pair · gpt-3.5-turbo

    the word "which" as it relates to clauses providing additional information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Behind
    -0.72
    athi
    -0.70
    Bas
    -0.68
     Ott
    -0.64
    grim
    -0.62
    STE
    -0.62
     Showdown
    -0.61
     Buc
    -0.60
     Passage
    -0.60
    UG
    -0.60
    POSITIVE LOGITS
     resulted
    0.86
    soever
    0.85
    allows
    0.80
     admittedly
    0.80
     brings
    0.79
     incidentally
    0.79
     includes
    0.79
     culminated
    0.78
    milo
    0.77
     consists
    0.77
    Act Density 0.134%

    No Known Activations