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

    Neuronpedia

    INDEX
    Explanations

    statements containing conditional statements

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    icing
    -0.79
    cookie
    -0.70
    iced
    -0.66
    entials
    -0.65
    ppa
    -0.64
    emouth
    -0.63
    embed
    -0.62
    cedented
    -0.62
    htaking
    -0.62
    mega
    -0.62
    POSITIVE LOGITS
     forfeit
    1.06
     goto
    1.04
     revert
    0.93
    throw
    0.93
     blame
    0.88
     discard
    0.85
     abort
    0.85
     forfe
    0.85
    merce
    0.84
     disqual
    0.82
    Act Density 0.406%

    No Known Activations