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

    Neuronpedia

    INDEX
    Explanations

    words related to inherent characteristics or properties

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to inherent traits or qualities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    kers
    -0.95
    pherd
    -0.74
    eday
    -0.71
    uku
    -0.70
    rooms
    -0.69
    fare
    -0.68
     Psychiat
    -0.68
    cow
    -0.67
    annis
    -0.66
     Soda
    -0.66
    POSITIVE LOGITS
     cumbers
    0.96
    ulner
    0.87
    ities
    0.85
     weaknesses
    0.84
     dangers
    0.82
     flaws
    0.82
     inherent
    0.82
    urities
    0.82
    idad
    0.77
     qualities
    0.77
    Act Density 0.019%

    No Known Activations