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

    Neuronpedia

    INDEX
    Explanations

    phrases related to maximizing benefits or opportunities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.26
    1:0.08
    2:0.02
    3:0.10
    4:0.05
    5:0.13
    6:0.04
    7:0.03
    8:0.09
    9:0.11
    10:0.02
    11:0.02
    Negative Logits
     torn
    -1.75
     abruptly
    -1.70
     strangely
    -1.68
    uddenly
    -1.68
    inally
    -1.66
     suddenly
    -1.66
     mysteriously
    -1.64
     policemen
    -1.63
    etheless
    -1.61
     rusty
    -1.59
    POSITIVE LOGITS
     advantage
    2.41
     profits
    2.40
    ongevity
    2.35
     profitability
    2.27
    iability
    2.26
     maximize
    2.22
    osity
    2.16
     efficiency
    2.13
    imum
    2.12
     profit
    2.11
    Act Density 0.002%

    No Known Activations