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

    Neuronpedia

    INDEX
    Explanations

    specific times and locations mentioned in the text

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     advoc
    -0.66
    isphere
    -0.60
    Connector
    -0.58
     informants
    -0.58
     prescriptions
    -0.58
     straight
    -0.58
     offending
    -0.56
    casting
    -0.56
    selves
    -0.55
    biased
    -0.55
    POSITIVE LOGITS
    mosp
    1.07
    las
    1.01
    hens
    0.95
    onement
    0.93
     least
    0.93
    letico
    0.90
    mega
    0.87
    abase
    0.84
    yp
    0.83
    raz
    0.83
    Act Density 0.035%

    No Known Activations