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

    Neuronpedia

    INDEX
    Explanations

    quantities

    np_max-act · gemini-2.0-flash

    This neuron activates on the adverb “relatively,” especially when it’s used to qualify measurements (e.g., “relatively low,” “relatively high,” “relatively small”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    achine
    -0.06
    routine
    -0.06
    OTH
    -0.06
    rid
    -0.06
    .players
    -0.06
    _Statics
    -0.06
    iler
    -0.06
    vided
    -0.06
    INCLUDE
    -0.06
    που
    -0.06
    POSITIVE LOGITS
     Bake
    0.06
     stopwatch
    0.06
     chor
    0.06
     Harmony
    0.06
    structures
    0.06
     enzyme
    0.06
     чемпіон
    0.06
     hukuk
    0.06
    shot
    0.06
     mum
    0.06
    Act Density 0.028%

    No Known Activations