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

    Neuronpedia

    INDEX
    Explanations

    phrases related to quantitative descriptions and control

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    rique
    -0.71
    erva
    -0.69
    hello
    -0.66
     Mour
    -0.66
    ren
    -0.65
     corps
    -0.64
     Finish
    -0.64
    ops
    -0.64
    endon
    -0.63
    ando
    -0.63
    POSITIVE LOGITS
     fluct
    1.38
     determined
    1.29
     determines
    1.27
     proportional
    1.26
     variable
    1.20
     dependent
    1.20
     dictated
    1.20
     adjustable
    1.20
     varies
    1.16
     vary
    1.16
    Act Density 0.402%

    No Known Activations