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 detects API query parameter names (e.g. “ci”, “lac”, “mnc”) in the documentation.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .Connect
    -0.07
     ders
    -0.06
    chk
    -0.06
     والت
    -0.06
     GEN
    -0.06
    Fragment
    -0.06
     thriller
    -0.06
    альних
    -0.06
    -items
    -0.06
     prost
    -0.06
    POSITIVE LOGITS
    (()
    0.07
    	hit
    0.07
    _option
    0.06
    ड
    0.06
    排
    0.06
    /load
    0.06
    orry
    0.06
    PNG
    0.06
     根
    0.06
    这里
    0.06
    Act Density 0.003%

    No Known Activations