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

    Neuronpedia

    INDEX
    Explanations

    large numeric values occurring in close vicinity

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of numeric values, particularly large numbers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     heights
    -0.70
     newsp
    -0.64
     curls
    -0.57
     riches
    -0.57
     enthus
    -0.57
     dogma
    -0.57
     advoc
    -0.57
     sights
    -0.56
     omn
    -0.56
    brainer
    -0.56
    POSITIVE LOGITS
    000
    1.80
    500
    1.40
    700
    1.30
    800
    1.30
    600
    1.28
    900
    1.24
    400
    1.23
    300
    1.16
    200
    1.16
    820
    1.13
    Act Density 0.064%

    No Known Activations