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

    Neuronpedia

    INDEX
    Explanations

    words related to potential, threat, or risk

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating potentiality or possible outcomes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    baugh
    -0.84
    tein
    -0.82
    ger
    -0.80
    bowl
    -0.73
    shire
    -0.71
    ters
    -0.70
    gio
    -0.70
    bern
    -0.70
     Plate
    -0.69
    gers
    -0.69
    POSITIVE LOGITS
     jeopard
    0.93
     hazardous
    0.86
     synerg
    0.85
     contam
    0.85
     disrupt
    0.83
     lethal
    0.79
     conce
    0.78
     habitable
    0.77
     conclud
    0.77
     avert
    0.76
    Act Density 0.016%

    No Known Activations