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

    Neuronpedia

    INDEX
    Explanations

    comparisons

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens, especially floating‐point literals (e.g., decimal numbers) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Hệ
    -0.07
    628
    -0.07
    يلاد
    -0.07
     суду
    -0.07
     Saul
    -0.07
    GINE
    -0.07
     Jung
    -0.07
    .ut
    -0.06
     Wash
    -0.06
     ranger
    -0.06
    POSITIVE LOGITS
    ruise
    0.06
    _traits
    0.06
    #"
    0.06
    issen
    0.06
     deepen
    0.06
    .Interface
    0.06
    _STATE
    0.06
    	sort
    0.05
    lv
    0.05
    vanized
    0.05
    Act Density 0.024%

    No Known Activations