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

    Neuronpedia

    INDEX
    Explanations

    document

    np_max-act · gemini-2.0-flash

    The neuron activates on words used in giving technical instructions for linking or embedding content (e.g. “link,” “paste,” “embed,” “HTML,” “email,” “IM,” “website,” “document”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ök
    -0.07
    ₀
    -0.06
    čer
    -0.06
    十
    -0.06
    수를
    -0.06
                        ↵                    ↵
    -0.06
    زيز
    -0.06
    ایر
    -0.06
     формы
    -0.06
     rostlin
    -0.06
    POSITIVE LOGITS
    "The
    0.07
    Incre
    0.07
    "She
    0.07
     Permanent
    0.07
    >/<
    0.06
    istinguish
    0.06
    "'
    0.06
    /*↵
    0.06
    "He
    0.06
    }

↵
    0.06
    Act Density 0.003%

    No Known Activations