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

    Neuronpedia

    INDEX
    Explanations

    situations where the word "somehow" is used to express an unexpected outcome

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     reluct
    -1.01
     depic
    -1.01
     accla
    -0.98
     strick
    -0.95
     shenan
    -0.93
     suscep
    -0.92
     affor
    -0.90
     increa
    -0.90
     uninten
    -0.88
     Pamph
    -0.88
    POSITIVE LOGITS
     somehow
    0.95
    <bos>
    0.79
     Somehow
    0.75
    Somehow
    0.70
     GeoNames
    0.69
    AsUp
    0.68
     jsPsych
    0.65
    mistak
    0.64
    TintMode
    0.62
    culously
    0.61
    Act Density 0.105%

    No Known Activations