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

    Neuronpedia

    INDEX
    Explanations

    code/data

    np_max-act · gemini-2.0-flash

    The neuron fires on identifiers that name auxiliary context or state‐carrying parameters (e.g. session IDs, userData/meta/pointer fields) in API definitions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Dim
    -0.07
    ak
    -0.07
    negative
    -0.06
     noise
    -0.06
     transforms
    -0.06
    _p
    -0.06
    وپ
    -0.06
     luck
    -0.06
     number
    -0.06
     VIEW
    -0.06
    POSITIVE LOGITS
    	fr
    0.07
    ateř
    0.07
     PdfP
    0.06
     supreme
    0.06
     pottery
    0.06
    .middle
    0.06
     enr
    0.06
    .isValid
    0.06
     दल
    0.06
     recep
    0.06
    Act Density 0.108%

    No Known Activations