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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "stunt"

    oai_token-act-pair · gpt-3.5-turbo

    references to stunts and pranks in promotional or dramatic contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    oice
    -0.72
    ishops
    -0.71
    vironment
    -0.69
    elong
    -0.67
    cing
    -0.66
    oan
    -0.64
    uni
    -0.62
    lected
    -0.61
    ecause
    -0.61
    icrobial
    -0.61
    POSITIVE LOGITS
     stunt
    0.95
     stunts
    0.91
    hered
    0.75
    woman
    0.73
    smanship
    0.73
    strip
    0.73
    monkey
    0.71
     sled
    0.71
    olicy
    0.71
    crew
    0.70
    Act Density 0.010%

    No Known Activations