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

    Neuronpedia

    INDEX
    Explanations

    years

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens forming years or dates (e.g. 4-digit years and parts of date ranges).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     cest
    -0.07
    اعي
    -0.07
     другого
    -0.06
     بس
    -0.06
    HTTP
    -0.06
    ző
    -0.06
     бути
    -0.06
    noxious
    -0.06
    .reporting
    -0.06
    效果
    -0.06
    POSITIVE LOGITS
     Relative
    0.07
     �
    0.06
    -ignore
    0.06
     assumes
    0.06
     carriers
    0.06
     indexOf
    0.06
    elder
    0.06
     intrigue
    0.06
     wounded
    0.06
     reaff
    0.06
    Act Density 0.031%

    No Known Activations