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

    Neuronpedia

    INDEX
    Explanations

    phrases describing physical actions or capabilities

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate human achievements or capabilities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    externalActionCode
    -0.72
    adata
    -0.70
    Jordan
    -0.69
    Enlarge
    -0.67
    DragonMagazine
    -0.67
    Jews
    -0.66
    jon
    -0.63
    Cruz
    -0.63
    fml
    -0.62
    å°Ĩ
    -0.62
    POSITIVE LOGITS
     these
    0.86
     activity
    0.71
     succumb
    0.69
    these
    0.68
     pursuits
    0.66
    ividual
    0.66
     endeavors
    0.65
     destructive
    0.65
     exert
    0.64
     activities
    0.64
    Act Density 1.309%

    No Known Activations