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

    Neuronpedia

    INDEX
    Explanations

    exponents

    np_max-act · gemini-2.0-flash

    The neuron responds to isolated single-letter variable names (e.g. “r”, “z”, “n”, etc.) in math‐style expressions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    :</
    -0.08
     ted
    -0.07
     interval
    -0.06
     IntPtr
    -0.06
    .cd
    -0.06
     ander
    -0.06
    Goal
    -0.06
     soar
    -0.06
    _centers
    -0.06
    _pts
    -0.06
    POSITIVE LOGITS
    (guid
    0.08
     Marker
    0.07
     �
    0.06
     Introduced
    0.06
     Installing
    0.06
     competitiveness
    0.06
     провести
    0.06
    Detach
    0.06
    Resize
    0.06
     ultrasound
    0.06
    Act Density 0.001%

    No Known Activations