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

    Neuronpedia

    INDEX
    Explanations

    large quantities or significant numerical figures

    oai_token-act-pair · gpt-3.5-turbo

    instances of high numerical values or quantities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    NetMessage
    -0.83
    relations
    -0.77
    emis
    -0.68
    testers
    -0.66
    pei
    -0.65
    na
    -0.64
    handler
    -0.64
     interacted
    -0.64
    nan
    -0.64
    icer
    -0.63
    POSITIVE LOGITS
     proportions
    0.90
     amount
    0.84
     amounts
    0.83
     staggering
    0.71
     volumes
    0.71
     unbeliev
    0.70
     heights
    0.70
    Amount
    0.70
     sums
    0.69
     totals
    0.68
    Act Density 0.036%

    No Known Activations