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

    Neuronpedia

    INDEX
    Explanations

    numbered lists

    np_max-act · gemini-2.0-flash

    This neuron detects numbered list or section labels—especially the token “1” used to mark the first item or heading in enumerations.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Items
    -0.07
    frog
    -0.06
    cepts
    -0.06
     moeten
    -0.06
     tempfile
    -0.06
     symbol
    -0.06
    (make
    -0.06
    ﾄ
    -0.06
     ObservableCollection
    -0.06
    .iteritems
    -0.06
    POSITIVE LOGITS
     Oliv
    0.07
    reve
    0.07
     význam
    0.07
    면서
    0.07
     denies
    0.06
     전쟁
    0.06
     Ruiz
    0.06
    embre
    0.06
    الص
    0.06
     russ
    0.06
    Act Density 0.002%

    No Known Activations