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

    Neuronpedia

    INDEX
    Explanations

    quotation marks at the beginning or end of phrases

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
    <bos>
    -1.72
    /*!
    -0.77
     sog
    -0.69
    /***
    -0.61
     Stä
    -0.59
    Referencoj
    -0.59
     Ufer
    -0.57
     dras
    -0.57
     aggres
    -0.57
     nark
    -0.55
    POSITIVE LOGITS
     nmax
    0.97
     Joaqu
    0.78
     unspeak
    0.76
     😭😭
    0.73
     ingrat
    0.69
     ajustable
    0.68
     impra
    0.68
     Mérida
    0.68
     ados
    0.68
     Cadiz
    0.67
    Act Density 0.341%

    No Known Activations