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

    Neuronpedia

    INDEX
    Explanations

    unimportant or passing remarks

    oai_token-act-pair · gpt-3.5-turbo

    sentences that end with a semicolon

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    orts
    -0.87
    din
    -0.79
     spons
    -0.78
    izons
    -0.73
    senal
    -0.72
     resil
    -0.71
     helicop
    -0.70
    wing
    -0.70
    hedral
    -0.69
    ovy
    -0.69
    POSITIVE LOGITS
    -)
    1.20
     namely
    1.14
     whereas
    0.90
    âĢ¢âĢ¢
    0.89
     furthermore
    0.83
     Whereas
    0.82
    âĢ¢âĢ¢âĢ¢âĢ¢
    0.81
     moreover
    0.79
    }"
    0.78
     thence
    0.77
    Act Density 0.054%

    No Known Activations