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

    Neuronpedia

    INDEX
    Explanations

    negative responses or denials in conversational contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    enko
    -0.16
    idis
    -0.15
    him
    -0.15
     Euras
    -0.15
    acs
    -0.14
    instead
    -0.14
    etc
    -0.14
    Ok
    -0.14
    ÂŃing
    -0.14
     untranslated
    -0.14
    POSITIVE LOGITS
     absolutely
    0.26
     partly
    0.21
     partially
    0.21
     actually
    0.21
     Absolutely
    0.20
     sir
    0.20
    Absolutely
    0.19
     definitely
    0.18
    !
    0.18
     entirely
    0.18
    Act Density 0.160%

    No Known Activations