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

    Neuronpedia

    INDEX
    Explanations

    phrases related to news updates and announcements

    oai_token-act-pair · gpt-3.5-turbo

    frequent references to time or regular updates

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Conservative
    -0.68
    minster
    -0.60
    Government
    -0.59
    Newsletter
    -0.59
    Muslim
    -0.59
    racial
    -0.58
    efe
    -0.58
    Obama
    -0.58
    Econom
    -0.57
    ioned
    -0.57
    POSITIVE LOGITS
     congr
    0.99
     please
    0.97
     congratulations
    0.90
     we
    0.87
     suffice
    0.83
     Leilan
    0.82
     however
    0.79
    please
    0.78
     beware
    0.77
     THANK
    0.76
    Act Density 0.369%

    No Known Activations