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

    Neuronpedia

    INDEX
    Explanations

    through

    np_max-act · gemini-2.0-flash

    This neuron lights up on words containing doubled consonants (e.g. the “gg” in “aggregative” or the “nn” in “dunnage”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     reproduce
    -0.07
    лятор
    -0.07
     enthusi
    -0.07
     spiritual
    -0.07
     podrob
    -0.06
     dosud
    -0.06
    father
    -0.06
     Lovely
    -0.06
    Decoder
    -0.06
    vell
    -0.06
    POSITIVE LOGITS
    ::|
    0.06
    -rays
    0.06
    시
    0.06
    케이
    0.06
    Composition
    0.06
     &$
    0.06
    Girls
    0.06
    -largest
    0.06
     converge
    0.06
     filtered
    0.06
    Act Density 0.270%

    No Known Activations