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

    Neuronpedia

    INDEX
    Explanations

    phrases related to societal roles and responsibilities

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     reperto
    -0.66
    jad
    -0.63
    ":["
    -0.62
    ulent
    -0.61
    raints
    -0.60
    tsy
    -0.58
    ificant
    -0.58
    isons
    -0.58
    uristic
    -0.58
    igue
    -0.58
    POSITIVE LOGITS
     temporarily
    0.62
     avoiding
    0.61
     oneself
    0.61
     protecting
    0.59
     owning
    0.57
     appointing
    0.57
     eliminating
    0.56
     ensuring
    0.56
     preserving
    0.55
     confirming
    0.55
    Act Density 10.515%

    No Known Activations