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

    Neuronpedia

    INDEX
    Explanations

    words indicating contrast or opposition in contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Efq
    -0.93
     pinulongan
    -0.89
     Pelop
    -0.78
     IAEA
    -0.77
     Bolshe
    -0.77
    ſelf
    -0.77
     faſt
    -0.77
     hierogly
    -0.77
     itſelf
    -0.76
     houſe
    -0.76
    POSITIVE LOGITS
     the
    0.96
     it
    0.70
     this
    0.68
    The
    0.67
     
    0.67
     these
    0.64
     if
    0.64
    由于
    0.63
     The
    0.63
    Although
    0.62
    Act Density 0.498%

    No Known Activations