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

    Neuronpedia

    INDEX
    Explanations

    triangle, upper

    np_max-act · gemini-2.0-flash

    This neuron fires on mentions of the Cholesky factorization—i.e. tokens from “Cholesky” together with references to the lower-triangular L factor (words like “lower,” “triangular,” “L1,” “L2,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    terms related to matrix factorization techniques.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     fren
    -0.07
    arket
    -0.07
     tk
    -0.07
     teeth
    -0.06
     talk
    -0.06
    sequelize
    -0.06
     Sevent
    -0.06
     NEVER
    -0.06
    ighth
    -0.06
    igate
    -0.06
    POSITIVE LOGITS
    うん
    0.07
    (_('
    0.06
    �
    0.06
     �
    0.06
    .".
    0.06
    (Route
    0.06
    궁
    0.06
    ––
    0.05
     секрет
    0.05
     Initi
    0.05
    Act Density 0.003%

    No Known Activations