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

    Neuronpedia

    INDEX
    Explanations

    -

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens (digits, multi‐digit numbers, and numeric sequences such as “150,” “119,” “638,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     nutrition
    -0.07
    _relative
    -0.06
    उ
    -0.06
    polation
    -0.06
     Nice
    -0.06
    .sat
    -0.06
     セ
    -0.06
    _CAM
    -0.06
    (found
    -0.06
    ُس
    -0.06
    POSITIVE LOGITS
     التج
    0.07
    غط
    0.06
     simplement
    0.06
     تخصص
    0.06
    �
    0.06
    -li
    0.06
    .Generated
    0.06
    �
    0.06
    贫
    0.06
    otechnology
    0.06
    Act Density 0.002%

    No Known Activations