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

    Neuronpedia

    INDEX
    Explanations

    expressions of belief or conviction

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     lot
    -0.67
    ddy
    -0.58
    ък
    -0.57
     N
    -0.56
     sosi
    -0.56
     Sh
    -0.56
    А
    -0.56
    sh
    -0.55
     lots
    -0.55
    gil
    -0.55
    POSITIVE LOGITS
     believe
    3.36
    believe
    3.09
     BELIEVE
    2.96
     Believe
    2.94
     believes
    2.87
    Believe
    2.79
     believed
    2.68
     believing
    2.62
     belief
    2.51
    belie
    2.49
    Act Density 0.064%

    No Known Activations