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

    Neuronpedia

    INDEX
    Explanations

    2

    np_max-act · gemini-2.0-flash

    The neuron activates on table‐layout tokens and HTML-like table attributes (e.g. rowspan, colspan, style) indicating a table structure.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    olv
    -0.07
     disconnected
    -0.06
    664
    -0.06
     yy
    -0.06
    ूर
    -0.06
     Milwaukee
    -0.06
     EST
    -0.06
     constituency
    -0.06
    setw
    -0.06
     خوان
    -0.06
    POSITIVE LOGITS
     stip
    0.06
     IPPROTO
    0.06
    unistd
    0.06
     بده
    0.06
    .formData
    0.06
     Plaint
    0.06
    ，“
    0.06
     cigaret
    0.06
    (android
    0.06
    _MI
    0.06
    Act Density 0.001%

    No Known Activations