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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating negation or lack of certainty

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     lele
    -0.88
     fua
    -0.82
     Ibidem
    -0.81
     hina
    -0.80
     Simult
    -0.80
     pama
    -0.77
     Membre
    -0.77
     Amerik
    -0.76
     kram
    -0.76
     kasa
    -0.75
    POSITIVE LOGITS
     necessarily
    0.74
     may
    0.68
     siquiera
    0.59
     be
    0.58
    principalTable
    0.58
     might
    0.58
    <bos>
    0.56
    may
    0.55
     not
    0.55
    necessarily
    0.52
    Act Density 0.080%

    No Known Activations