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

    Neuronpedia

    INDEX
    Explanations

    numerical figures expressed as percentages or quantities

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    awar
    -0.59
     resemblance
    -0.58
    lessly
    -0.58
     therm
    -0.57
     recoil
    -0.57
    emon
    -0.56
    "))
    -0.56
     ¯
    -0.53
     vacuum
    -0.53
    quickShipAvailable
    -0.53
    POSITIVE LOGITS
    bps
    0.74
    %.
    0.71
     manageable
    0.69
    kus
    0.68
    livion
    0.67
    éĹĺ
    0.65
    ãĥ¯
    0.64
     digits
    0.64
     destinations
    0.62
    nr
    0.62
    Act Density 0.170%

    No Known Activations