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

    Neuronpedia

    INDEX
    Explanations

    references to specific numerical data or metadata related to research articles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    URLs and technical identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused28>
    -0.93
    <unused68>
    -0.93
    <unused41>
    -0.93
    <unused8>
    -0.93
    <unused79>
    -0.93
    [@BOS@]
    -0.93
    <unused14>
    -0.93
    <unused52>
    -0.93
    <unused16>
    -0.93
    <unused23>
    -0.93
    POSITIVE LOGITS
     indicado
    0.35
    ),
    0.30
     claro
    0.29
    .
    0.28
    depuis
    0.26
     quizá
    0.26
    ;
    0.26
    );
    0.26
    ↵
    0.26
    ViewImports
    0.25
    Act Density 0.018%

    No Known Activations