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

    Neuronpedia

    INDEX
    Explanations

    detailed descriptions about locations and events

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    conom
    -0.89
     predictive
    -0.80
    binding
    -0.77
     seeks
    -0.76
     epigen
    -0.75
     endeavors
    -0.73
     analogous
    -0.71
     embodies
    -0.71
    ingen
    -0.70
     immortality
    -0.70
    POSITIVE LOGITS
     crews
    1.10
     luckily
    1.00
     passers
    0.98
     guests
    0.92
     passengers
    0.91
     troopers
    0.91
     occupants
    0.91
     policemen
    0.90
    Luckily
    0.90
     paramedics
    0.89
    Act Density 0.901%

    No Known Activations