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

    Neuronpedia

    INDEX
    Explanations

    references to deception and falsehoods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     surla
    -0.68
     calendriers
    -0.65
    #+#
    -0.58
    uintptr
    -0.57
     AssemblyProduct
    -0.57
     CGRectMake
    -0.57
    unknownFields
    -0.57
    UObject
    -0.55
    bigsqcup
    -0.55
    CGRectMake
    -0.54
    POSITIVE LOGITS
     lied
    1.03
     lie
    0.96
     lies
    0.94
     liars
    0.93
     liar
    0.93
     falsehood
    0.92
    Lies
    0.90
     deceitful
    0.90
     Lying
    0.88
     deception
    0.88
    Act Density 0.443%

    No Known Activations