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

    Neuronpedia

    INDEX
    Explanations

    the word "sake" which usually appears in phrases indicating reasons or purposes. It may be looking for intentions or justifications

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "for the sake of" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ibr
    -0.67
    orial
    -0.64
    eval
    -0.64
    RC
    -0.63
     Radar
    -0.62
    line
    -0.60
    BM
    -0.55
    ve
    -0.54
    Lock
    -0.54
    Scan
    -0.54
    POSITIVE LOGITS
     sake
    4.34
     purposes
    1.53
     purpose
    1.06
     convenience
    1.01
     reasons
    0.97
    theless
    0.88
     simplicity
    0.88
     detriment
    0.86
    teness
    0.82
     entirety
    0.80
    Act Density 0.016%

    No Known Activations