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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "weakness" along with the context in which it appears

    oai_token-act-pair · gpt-3.5-turbo

    references to weaknesses in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     informed
    -0.75
     notified
    -0.70
     escorted
    -0.69
     billed
    -0.68
     seamlessly
    -0.65
     decorated
    -0.65
     Electronic
    -0.63
    cop
    -0.63
     Dot
    -0.63
    cend
    -0.63
    POSITIVE LOGITS
     weakness
    3.98
     weaknesses
    2.80
     Weak
    2.22
    Weak
    2.14
    weak
    1.89
     weak
    1.86
     weakening
    1.77
     vulnerability
    1.76
     strength
    1.71
     weakest
    1.66
    Act Density 0.021%

    No Known Activations