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

    Neuronpedia

    INDEX
    Explanations

    multiple instances of the word "comment" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ialis
    -0.16
    786
    -0.16
    worm
    -0.15
    758
    -0.14
    got
    -0.14
     Bloom
    -0.14
    lom
    -0.14
    chet
    -0.14
    ãĥ³ãĥĶ
    -0.14
    endale
    -0.14
    POSITIVE LOGITS
    (#)
    0.16
    ghan
    0.15
    imbus
    0.15
    aries
    0.15
    /Instruction
    0.14
    eded
    0.14
    _inches
    0.14
    湿
    0.14
    ç³»åĪĹ
    0.14
     persu
    0.14
    Act Density 0.023%

    No Known Activations