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

    Neuronpedia

    INDEX
    Explanations

    ="

    np_max-act · gemini-2.0-flash

    This neuron detects numeric tokens with decimal points (floating-point numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     initials
    -0.06
     themed
    -0.06
    626
    -0.06
     monarchy
    -0.06
    672
    -0.06
     CGSizeMake
    -0.06
    iliar
    -0.06
    邊
    -0.06
     závě
    -0.06
     Vương
    -0.06
    POSITIVE LOGITS
    	Random
    0.07
     повер
    0.07
    /Open
    0.07
    este
    0.07
     наче
    0.06
    /std
    0.06
     bat
    0.06
     compiled
    0.06
     �
    0.06
    στ
    0.06
    Act Density 0.003%

    No Known Activations