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

    Neuronpedia

    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    This neuron detects date-related tokens (days of the week, months, day numbers, commas, and years) in document headings.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    anik
    -0.06
     Warm
    -0.06
    -west
    -0.06
    regar
    -0.06
     Humans
    -0.06
     Alarm
    -0.06
    ışman
    -0.06
    ран
    -0.06
     Cher
    -0.06
    xFFFF
    -0.06
    POSITIVE LOGITS
     низ
    0.07
    _scene
    0.07
     coraz
    0.06
     worked
    0.06
    (forKey
    0.06
     totaling
    0.06
     bronze
    0.06
     genetically
    0.06
     knife
    0.06
    Canon
    0.06
    Act Density 0.005%

    No Known Activations