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

    Neuronpedia

    INDEX
    Explanations

    properties

    np_max-act · gemini-2.0-flash

    This neuron detects the presence of JSON Schema structure keywords (like “properties,” “type,” “required,” etc.) indicating a JSON Schema definition.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    dir
    -0.07
    Cancel
    -0.07
    Scan
    -0.06
     sopr
    -0.06
     darkest
    -0.06
     Flower
    -0.06
    Difficulty
    -0.06
    过
    -0.06
    Indexed
    -0.06
    /tutorial
    -0.06
    POSITIVE LOGITS
    aryawan
    0.06
     Güvenlik
    0.06
     cardboard
    0.06
     Bever
    0.06
    ίκ
    0.06
     Six
    0.06
    urance
    0.06
    ITES
    0.06
    εργ
    0.06
    enity
    0.06
    Act Density 0.001%

    No Known Activations