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

    Neuronpedia

    INDEX
    Explanations

    capital letters in an unusual format

    oai_token-act-pair · gpt-3.5-turbo

    specific symbol patterns and certain unique characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ithing
    -0.75
     Peb
    -0.69
    ulators
    -0.68
    bro
    -0.67
    iations
    -0.64
     Roads
    -0.63
    aturdays
    -0.63
    theless
    -0.62
    oys
    -0.62
    MEN
    -0.61
    POSITIVE LOGITS
    ª
    1.49
    ©
    1.44
    Ļ
    1.38
    Ķ
    1.31
    ¢
    1.31
    ¾
    1.29
    ¡
    1.28
    ¨
    1.27
    IJ
    1.24
    Ľ
    1.24
    Act Density 0.028%

    No Known Activations