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

    Neuronpedia

    INDEX
    Explanations

    the mention or discussion of opponents in a strategic or competitive context

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    olia
    -0.73
    oots
    -0.66
    mberg
    -0.66
     Loch
    -0.65
    ulic
    -0.62
    vest
    -0.62
    cow
    -0.61
    urgical
    -0.61
    erd
    -0.60
    umph
    -0.60
    POSITIVE LOGITS
     batters
    0.85
    onent
    0.80
     opponent
    0.77
     foe
    0.72
     combatants
    0.67
    hesis
    0.66
    dayName
    0.64
     opponents
    0.63
     vanquished
    0.63
    oft
    0.60
    Act Density 5.838%

    No Known Activations