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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a focus on specific topics or concerns

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    dayName
    -0.65
    MpServer
    -0.64
    berus
    -0.64
    hya
    -0.60
     condem
    -0.60
     commute
    -0.60
     lapt
    -0.59
    oney
    -0.59
     disposed
    -0.58
     agre
    -0.58
    POSITIVE LOGITS
    arching
    0.77
    terness
    0.65
    icial
    0.64
     Decay
    0.63
     Lies
    0.62
     reasons
    0.61
     earliest
    0.61
    sers
    0.61
    arest
    0.60
    ounding
    0.60
    Act Density 0.087%

    No Known Activations