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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating expectation or lack thereof

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate a lack of surprise or unexpectedness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    nels
    -0.72
     Logged
    -0.67
    abulary
    -0.61
    audi
    -0.60
    nis
    -0.59
     Suns
    -0.59
    tery
    -0.59
     gou
    -0.58
    heastern
    -0.56
    ouf
    -0.56
    POSITIVE LOGITS
     surprise
    0.82
    actionDate
    0.73
     shock
    0.70
    ENSE
    0.69
    ream
    0.65
    rection
    0.64
    azes
    0.64
    ratulations
    0.63
     though
    0.63
    card
    0.62
    Act Density 0.068%

    No Known Activations