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

    Neuronpedia

    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    The neuron fires on spelled‐out numerals (e.g. “two,” “three”) used to enumerate methods, objectives, or groups.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (generator
    -0.07
    ARG
    -0.07
     disregard
    -0.07
    -heart
    -0.07
     elev
    -0.07
     undo
    -0.06
    	pass
    -0.06
     phone
    -0.06
    Pros
    -0.06
     отп
    -0.06
    POSITIVE LOGITS
     Proj
    0.07
     JsonResult
    0.07
    •↵↵
    0.07
    (',');↵
    0.06
    (JSONObject
    0.06
    */↵↵↵
    0.06
    另外
    0.06
    óln
    0.06
     <!--
    0.06
    [['
    0.06
    Act Density 0.032%

    No Known Activations