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

    Neuronpedia

    INDEX
    Explanations

    verbs that indicate making a conscious decision

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "choose" indicating decisions or choices being made

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    brance
    -0.78
    urst
    -0.70
    uum
    -0.69
    utenberg
    -0.67
    otor
    -0.66
    atomic
    -0.64
    ijk
    -0.64
    Net
    -0.63
    awar
    -0.61
    iter
    -0.61
    POSITIVE LOGITS
     chooses
    0.89
     chose
    0.89
     choose
    0.83
    axe
    0.81
     chosen
    0.78
     choices
    0.76
     ACTIONS
    0.76
     choosing
    0.71
     wisely
    0.70
    çĶŁ
    0.70
    Act Density 0.026%

    No Known Activations