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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron never fires on any of these code or text tokens—it appears to be effectively inactive (i.e. it doesn’t detect any specific pattern).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    BYTE
    -0.06
    _com
    -0.06
    _four
    -0.06
    -panel
    -0.06
     Fou
    -0.06
     эта
    -0.06
     tunnel
    -0.06
     dataArray
    -0.06
    ною
    -0.06
    �
    -0.06
    POSITIVE LOGITS
    irts
    0.07
     Owned
    0.07
    ="./
    0.07
    บาย
    0.06
     справи
    0.06
     podnikatel
    0.06
    [G
    0.06
    kelas
    0.06
     subscribe
    0.06
     competent
    0.06
    Act Density 0.002%

    No Known Activations