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

    Neuronpedia

    INDEX
    Explanations

    reduce

    np_max-act · gemini-2.0-flash

    The neuron fires on gerund forms of “reduce” (e.g. “reducing”) and related phrasing about cutting or lowering quantities—especially in the context of waste, resource use, or environmental impact.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     depart
    -0.08
     Вот
    -0.07
     обычно
    -0.07
     pasture
    -0.06
     �
    -0.06
     ситуации
    -0.06
     ONE
    -0.06
    进一步
    -0.06
     mà
    -0.06
     te
    -0.06
    POSITIVE LOGITS
     Plugins
    0.06
     Wool
    0.06
    .Observable
    0.06
    _dimension
    0.06
     Dec
    0.06
     Dive
    0.06
     SEC
    0.06
    filesystem
    0.06
    tě
    0.06
    _fl
    0.06
    Act Density 0.015%

    No Known Activations