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

    Neuronpedia

    INDEX
    Explanations

    phrases related to emergence and creation

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate relationships and connections among various entities and concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    thing
    -0.78
    he
    -0.75
    LC
    -0.72
    cles
    -0.70
    iture
    -0.69
    rency
    -0.69
    hes
    -0.68
    ilty
    -0.67
    soever
    -0.67
    ire
    -0.66
    POSITIVE LOGITS
     several
    0.88
     numerous
    0.83
     what
    0.74
     countless
    0.72
     locating
    0.71
     myriad
    0.71
     innumerable
    0.70
     both
    0.70
     various
    0.70
     those
    0.68
    Act Density 0.722%

    No Known Activations