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

    Neuronpedia

    INDEX
    Explanations

    answer

    np_max-act · gemini-2.0-flash

    This neuron detects Q&A metadata terms—words referring to posts and their actions (e.g. “question,” “answer,” “comment,” “wiki,” “post”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    croll
    -0.07
     commanding
    -0.07
    -circle
    -0.07
    Mission
    -0.07
     스트
    -0.06
     geil
    -0.06
     Lease
    -0.06
     CircularProgress
    -0.06
     ASIC
    -0.06
    獎
    -0.06
    POSITIVE LOGITS
    (pointer
    0.06
     linea
    0.06
    аліз
    0.06
    ">×</
    0.06
    _STATIC
    0.05
     grav
    0.05
     eagerly
    0.05
     pov
    0.05
    (...)↵
    0.05
    امة
    0.05
    Act Density 0.004%

    No Known Activations