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

    Neuronpedia

    INDEX
    Explanations

    conditional phrases and expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    roman
    -0.16
    /interface
    -0.15
    .mixin
    -0.14
     Kraj
    -0.14
    uela
    -0.14
     تÙĤس
    -0.14
    eel
    -0.13
    lug
    -0.13
    =__
    -0.13
    uj
    -0.13
    POSITIVE LOGITS
    rames
    0.16
    /how
    0.15
     Mun
    0.15
    obox
    0.15
     necessary
    0.15
    boxes
    0.14
    ĶĦ
    0.14
     correct
    0.14
     there
    0.14
    бÑĢа
    0.14
    Act Density 0.023%

    No Known Activations