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

    Neuronpedia

    INDEX
    Explanations

    phrases related to deletion or removal

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "taking down."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    inen
    -0.82
    ature
    -0.69
    ugh
    -0.65
    uno
    -0.65
    yne
    -0.65
     Opinion
    -0.65
    aternity
    -0.64
    athi
    -0.64
    cius
    -0.63
    nder
    -0.62
    POSITIVE LOGITS
     stairs
    1.06
    graded
    0.86
    hill
    0.86
    grading
    0.81
     tablets
    0.71
     monuments
    0.71
    LOAD
    0.71
    stairs
    0.70
     stakes
    0.70
     flyers
    0.69
    Act Density 0.031%

    No Known Activations