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

    Neuronpedia

    INDEX
    Explanations

    choices or decision-making tasks

    oai_token-act-pair · gpt-3.5-turbo

    references to making choices or decisions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Schwe
    -0.74
    ãĤ¸
    -0.72
    aud
    -0.69
    ¶ħ
    -0.66
    enium
    -0.66
    ©¶æ¥µ
    -0.65
    ulz
    -0.64
    TPS
    -0.64
    arag
    -0.64
     Mare
    -0.63
    POSITIVE LOGITS
     choices
    1.97
     choice
    1.95
    choice
    1.77
     choosing
    1.75
    Choice
    1.71
     Choice
    1.66
     choose
    1.64
     chose
    1.63
    Option
    1.57
     chooses
    1.53
    Act Density 0.608%

    No Known Activations