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

    Neuronpedia

    INDEX
    Explanations

    research paper introductions

    np_max-act · gemini-2.0-flash

    The neuron primarily detects floating‐point numeric tokens (e.g. “2.890625”) that appear in section headings or formatting.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _scope
    -0.08
    onym
    -0.08
     Mapper
    -0.07
    μένοι
    -0.06
     Pil
    -0.06
    уль
    -0.06
    	Main
    -0.06
     Oven
    -0.06
     jane
    -0.06
    Kit
    -0.06
    POSITIVE LOGITS
    (':',
    0.06
    _styles
    0.06
     impressive
    0.06
    ΥΝ
    0.06
    528
    0.06
     어
    0.06
     fourn
    0.06
     Çev
    0.06
    _voltage
    0.06
     bulky
    0.06
    Act Density 0.009%

    No Known Activations