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

    Neuronpedia

    INDEX
    Explanations

    the word "rather" and phrases related to preference or choice

    oai_token-act-pair · gpt-3.5-turbo

    expressions of preference or choices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rising
    -0.76
    calling
    -0.72
    oward
    -0.70
    emin
    -0.68
    apt
    -0.67
    Loading
    -0.63
    DAQ
    -0.62
    sembly
    -0.62
    essential
    -0.62
    ahead
    -0.60
    POSITIVE LOGITS
     spend
    1.08
     lose
    1.06
     starve
    1.03
     avoid
    0.98
     die
    0.95
     stay
    0.93
     gamble
    0.91
     rely
    0.89
     endure
    0.89
     be
    0.89
    Act Density 0.046%

    No Known Activations