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

    Neuronpedia

    INDEX
    Explanations

    years expressed in numerical format

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     ancestors
    -0.71
    é¾įå¥ij士
    -0.69
    intent
    -0.64
    :/
    -0.64
    eyes
    -0.58
    reads
    -0.58
    hyde
    -0.57
     constituents
    -0.57
    selves
    -0.57
    tx
    -0.56
    POSITIVE LOGITS
    sequently
    1.09
     1994
    1.05
     subsequent
    1.05
     thereafter
    1.05
     1993
    1.04
     1991
    1.03
     1992
    1.02
     1986
    1.02
    later
    1.02
     1990
    1.01
    Act Density 0.316%

    No Known Activations