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

    Neuronpedia

    INDEX
    Explanations

    punctuation marks and their associated labels in structured text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    fleisch
    -0.53
    uD
    -0.51
    inės
    -0.49
    aliyet
    -0.49
    väg
    -0.48
     kří
    -0.48
     of
    -0.47
    leſs
    -0.47
     הה
    -0.47
    gott
    -0.46
    POSITIVE LOGITS
    ",
    2.54
    )",
    2.28
    ”,
    2.15
    ?",
    2.15
    ]",
    2.06
    ,",
    2.04
    .",
    2.04
    !",
    2.01
    }",
    2.00
    :",
    1.98
    Act Density 0.118%

    No Known Activations