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

    Neuronpedia

    INDEX
    Explanations

    phrases that denote representation or acting on someone else's behalf

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     sap
    -0.48
    痴
    -0.48
    wira
    -0.46
    ishi
    -0.45
    τας
    -0.45
    劣
    -0.44
     Tre
    -0.43
     
    -0.42
    sap
    -0.42
     Gom
    -0.41
    POSITIVE LOGITS
     mewakili
    1.01
     representing
    0.98
     behalf
    0.94
     Representing
    0.90
    representing
    0.89
    IsMutable
    0.83
     Represent
    0.78
    wakili
    0.77
     houſe
    0.77
     perſon
    0.77
    Act Density 0.200%

    No Known Activations