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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a typical or expected outcome or behavior

    oai_token-act-pair · gpt-3.5-turbo

    words that indicate common characteristics or typical behaviors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     possibly
    -0.72
    amins
    -0.66
    hab
    -0.65
    ingly
    -0.65
    together
    -0.64
    iage
    -0.62
    DNA
    -0.62
    ————————
    -0.61
    yle
    -0.61
     potentially
    -0.60
    POSITIVE LOGITS
     reserved
    0.94
     speaking
    0.81
     refers
    0.75
     wont
    0.75
     suspects
    0.74
     associated
    0.73
     disclaim
    0.71
     consists
    0.71
     accompanies
    0.69
     regarded
    0.69
    Act Density 0.106%

    No Known Activations