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

    Neuronpedia

    INDEX
    Explanations

    reported events or incidents

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    hart
    -0.65
    lain
    -0.60
    Enhanced
    -0.59
    ors
    -0.57
    heid
    -0.55
    ainment
    -0.54
    holding
    -0.54
    bay
    -0.53
    istan
    -0.52
    ament
    -0.52
    POSITIVE LOGITS
     into
    0.80
     sideways
    0.80
     overboard
    0.76
     onto
    0.75
     seamlessly
    0.75
     downward
    0.72
     INTO
    0.71
     away
    0.71
     downwards
    0.70
     upward
    0.70
    Act Density 13.490%

    No Known Activations