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

    Neuronpedia

    INDEX
    Explanations

    Physics; Lagrangian formulas

    np_max-act · gemini-2.0-flash

    This neuron activates on the split‐up pieces of the word “Lagrangian,” effectively detecting occurrences of that term in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _pe
    -0.07
    .ref
    -0.07
     sturdy
    -0.07
     hostname
    -0.07
    realm
    -0.07
     Parsing
    -0.07
    .Lo
    -0.06
     real
    -0.06
    τη
    -0.06
    	header
    -0.06
    POSITIVE LOGITS
    ucken
    0.06
    clarsimp
    0.06
     Samar
    0.06
     Roz
    0.06
     antenn
    0.06
    тура
    0.06
     Participation
    0.06
    ладу
    0.06
     Renault
    0.06
    ypsum
    0.06
    Act Density 0.003%

    No Known Activations