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

    Neuronpedia

    INDEX
    Explanations

    handwritten text, signatures, and text written with a pen

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    nels
    -0.79
    places
    -0.78
    irth
    -0.69
    respect
    -0.66
     Gim
    -0.65
    tackle
    -0.65
     Robertson
    -0.64
    γ
    -0.64
    bringing
    -0.62
    owered
    -0.62
    POSITIVE LOGITS
     scrib
    0.93
    kered
    0.92
     notebook
    0.89
     detailing
    0.86
     scrolls
    0.84
     documents
    0.83
     handwriting
    0.83
     notebooks
    0.83
     memos
    0.81
     handwritten
    0.81
    Act Density 3.546%

    No Known Activations