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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a preference for a specific option or action

    oai_token-act-pair · gpt-3.5-turbo

    expressions of preference or choice

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rising
    -0.74
    emin
    -0.70
    uther
    -0.70
     afore
    -0.64
    apt
    -0.62
    calling
    -0.62
    anni
    -0.61
    Loading
    -0.60
    ahead
    -0.60
    Behind
    -0.60
    POSITIVE LOGITS
     lose
    0.96
     than
    0.96
     spend
    0.95
     avoid
    0.86
     starve
    0.85
     settle
    0.81
     gamble
    0.80
     tolerate
    0.79
     wait
    0.79
     accept
    0.78
    Act Density 0.042%

    No Known Activations