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

    Neuronpedia

    INDEX
    Explanations

    phrases related to staying informed or keeping updated

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to maintaining awareness or staying informed

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    BALL
    -0.60
    furt
    -0.60
    mology
    -0.59
    Rus
    -0.59
    atis
    -0.57
     Barg
    -0.57
    hiro
    -0.56
     Audi
    -0.56
     Fields
    -0.56
     Ravens
    -0.56
    POSITIVE LOGITS
     appearances
    0.93
     pace
    0.75
    ouver
    0.70
    earances
    0.70
    dates
    0.69
    ency
    0.68
    dating
    0.68
    standing
    0.67
    endra
    0.67
     footing
    0.65
    Act Density 0.030%

    No Known Activations