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

    Neuronpedia

    INDEX
    Explanations

    phrases or words related to debates, discussions or questions

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    uras
    -0.64
    amina
    -0.58
    hetical
    -0.58
    urances
    -0.58
    iaries
    -0.55
    ogly
    -0.54
    istically
    -0.54
    alty
    -0.53
    ionage
    -0.53
    urance
    -0.53
    POSITIVE LOGITS
     raging
    0.71
    naire
    0.70
    puff
    0.70
     raged
    0.69
    ably
    0.66
     moderators
    0.65
    halla
    0.64
     debate
    0.63
    ively
    0.61
    dayName
    0.60
    Act Density 8.125%

    No Known Activations