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

    Neuronpedia

    INDEX
    Explanations

    adjectives related to human qualities and behavior, such as bravery, stupidity, sincerity, and hypocrisy

    oai_token-act-pair · gpt-3.5-turbo

    characteristics related to personal qualities or societal critiques

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    romeda
    -0.69
    ãĤ´ãĥ³
    -0.68
    itionally
    -0.66
    iris
    -0.66
    tainment
    -0.64
    nesium
    -0.62
     ado
    -0.59
    amera
    -0.58
    ãĤ¡
    -0.57
    ourt
    -0.57
    POSITIVE LOGITS
     exhibited
    1.12
     inherent
    1.06
     displayed
    0.99
     evident
    0.95
    lessness
    0.92
     afforded
    0.91
     shown
    0.89
     exercised
    0.84
     of
    0.83
     required
    0.82
    Act Density 0.169%

    No Known Activations