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

    Neuronpedia

    INDEX
    Explanations

    words related to agreements or deals between entities

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "agree" along with its derivatives in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    grain
    -0.75
    patch
    -0.72
    clips
    -0.69
    pointer
    -0.68
     wastes
    -0.66
    ults
    -0.65
     Detected
    -0.64
    worm
    -0.64
    worms
    -0.62
    ï¸ı
    -0.61
    POSITIVE LOGITS
     waive
    1.06
     cooperate
    1.04
     accept
    1.03
     settle
    0.97
     participate
    0.96
     obey
    0.95
     submit
    0.95
     surrender
    0.94
     abide
    0.93
     undertake
    0.89
    Act Density 0.062%

    No Known Activations