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

    Neuronpedia

    INDEX
    Explanations

    phrases related to attributing responsibility or blame

    oai_token-act-pair · gpt-3.5-turbo

    references to accountability and blame in context of actions and consequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Tree
    -0.79
    isSpecialOrderable
    -0.78
    eki
    -0.74
    vine
    -0.72
    adj
    -0.69
    herer
    -0.69
    atories
    -0.68
    paces
    -0.66
     Bake
    -0.65
    ibaba
    -0.65
    POSITIVE LOGITS
     sins
    1.20
     sake
    1.13
     inconvenience
    1.08
     crimes
    0.98
     transgress
    0.95
     failings
    0.90
     failures
    0.88
     deaths
    0.87
     offences
    0.87
     mistakes
    0.86
    Act Density 0.326%

    No Known Activations