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

    Neuronpedia

    INDEX
    Explanations

    names of athletes and sports-related statistics

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     lull
    -0.75
     nic
    -0.74
     intellectual
    -0.71
     manufact
    -0.70
     shadows
    -0.70
     tenants
    -0.69
     creation
    -0.69
     creations
    -0.67
     dow
    -0.65
     publishing
    -0.65
    POSITIVE LOGITS
    albeit
    1.32
    six
    1.29
    seven
    1.27
    five
    1.26
    although
    1.24
    four
    1.24
    multiple
    1.23
    both
    1.23
    three
    1.22
    third
    1.21
    Act Density 0.095%

    No Known Activations