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

    Neuronpedia

    INDEX
    Explanations

    ,

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens associated with numbers, dates, counts, and other factual or historical numeric references in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (The
    -0.07
     upcoming
    -0.07
     ksi
    -0.06
    .Headers
    -0.06
     смот
    -0.06
     помощью
    -0.06
     deepest
    -0.06
    Parts
    -0.06
    _gui
    -0.06
    -blue
    -0.06
    POSITIVE LOGITS
    ersions
    0.06
    mj
    0.06
    `;
    0.06
    оди
    0.06
     �
    0.06
     emerg
    0.06
     √
    0.06
    mt
    0.06
    _Valid
    0.06
     INTERN
    0.06
    Act Density 0.143%

    No Known Activations