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

    Neuronpedia

    INDEX
    Explanations

    limit/restrict

    np_max-act · gemini-2.0-flash

    This neuron fires on expressions of quantitative limits or restrictions (e.g., “limited,” “limit,” “maximum,” “restriction”) in technical descriptions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ्पन
    -0.07
    ocusing
    -0.07
    come
    -0.06
     adipisicing
    -0.06
    .Messaging
    -0.06
    CommandLine
    -0.06
    non
    -0.06
    .orientation
    -0.06
    ouden
    -0.06
    MessageBox
    -0.06
    POSITIVE LOGITS
     medya
    0.06
     vậy
    0.06
     crest
    0.06
     행복
    0.06
     velik
    0.06
     deselect
    0.06
     lush
    0.06
     dễ
    0.06
    _reserved
    0.06
     ranger
    0.06
    Act Density 0.020%

    No Known Activations