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

    Neuronpedia

    INDEX
    Explanations

    numerical height measurements

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    utical
    -0.78
     Reloaded
    -0.75
    ño
    -0.73
    zsche
    -0.72
    tical
    -0.71
    BOOK
    -0.68
    ãģ¦
    -0.68
    ãĥ£
    -0.65
    hyde
    -0.63
    eer
    -0.63
    POSITIVE LOGITS
    ened
    1.01
    ening
    0.96
     stature
    0.82
    falls
    0.81
    rises
    0.80
     height
    0.80
     cliffs
    0.79
    ensions
    0.78
    band
    0.77
    map
    0.77
    Act Density 0.039%

    No Known Activations