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

    Neuronpedia

    INDEX
    Explanations

    forms of emphasis, such as exclamation marks or capitalized words, which indicate strong emotions or stress

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
    %%%%
    -0.73
     @@
    -0.65
     Scrolls
    -0.64
     Theft
    -0.59
    quickShipAvailable
    -0.59
     signs
    -0.56
     smokes
    -0.56
     splits
    -0.55
     crept
    -0.55
    missing
    -0.55
    POSITIVE LOGITS
    venge
    1.37
     extent
    1.36
     detriment
    1.29
     tune
    1.13
    rouse
    1.12
    livion
    1.03
     shores
    0.97
     fullest
    0.95
     liking
    0.90
     brink
    0.88
    Act Density 4.801%

    No Known Activations