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

    Neuronpedia

    INDEX
    Explanations

    phrases related to expectations or obligations not being met

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to expectations or obligations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iple
    -0.73
     Appears
    -0.68
    lip
    -0.65
    backer
    -0.61
     quotations
    -0.61
    Reader
    -0.61
    river
    -0.59
    ious
    -0.59
     Tens
    -0.59
    mons
    -0.59
    POSITIVE LOGITS
     behave
    0.92
     be
    0.90
     uphold
    0.85
     defend
    0.85
     abide
    0.83
     embody
    0.82
     represent
    0.82
     steer
    0.81
     compensate
    0.81
     adhere
    0.80
    Act Density 0.089%

    No Known Activations