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

    Neuronpedia

    INDEX
    Explanations

    instances where the subject is described as starting a physical action but is prevented from completing it

    oai_token-act-pair · gpt-3.5-turbo

    the conjunction "but" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ermanent
    -0.76
    arity
    -0.74
    onomy
    -0.69
    ability
    -0.67
    abilities
    -0.67
    RAW
    -0.67
    irt
    -0.66
    aring
    -0.64
    ands
    -0.63
    iola
    -0.63
    POSITIVE LOGITS
     alas
    1.04
     luckily
    1.01
     fortunately
    0.97
     withdrew
    0.96
    tons
    0.88
     thankfully
    0.88
    ersen
    0.87
     forgot
    0.86
     lacked
    0.85
     opted
    0.83
    Act Density 0.177%

    No Known Activations