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

    Neuronpedia

    INDEX
    Explanations

    structured data/document extractions

    np_max-act · gemini-2.0-flash

    The neuron is detecting special control/metadata tokens (e.g. header- and end-of-text markers) in the transcript.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     aden
    -0.06
    金属
    -0.06
    论文
    -0.06
    .Detail
    -0.06
    Multiplicity
    -0.06
    _next
    -0.06
    ideal
    -0.06
    ниці
    -0.06
    	nodes
    -0.06
    getc
    -0.05
    POSITIVE LOGITS
     interior
    0.07
    ewire
    0.07
     México
    0.07
     Harry
    0.07
    .Net
    0.07
     Richmond
    0.07
     PS
    0.07
     İngiliz
    0.07
    .net
    0.07
    -submit
    0.07
    Act Density 0.002%

    No Known Activations