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

    Neuronpedia

    INDEX
    Explanations

    figure

    np_max-act · gemini-2.0-flash

    This neuron specifically fires on in-text figure references (e.g. “Figure 1,” “[#fig…]”) and related tokens.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    GeneratedValue
    -0.07
     Как
    -0.07
     uterus
    -0.07
     Вот
    -0.06
     welded
    -0.06
    όμε
    -0.06
    ialized
    -0.06
    وه
    -0.06
     Undefined
    -0.06
    vrolet
    -0.06
    POSITIVE LOGITS
     showModal
    0.07
     editar
    0.06
     Feld
    0.06
    .gr
    0.06
     бит
    0.06
     transitioning
    0.06
    ैं.
    0.06
    ToInt
    0.06
    십시오
    0.06
     uri
    0.06
    Act Density 0.006%

    No Known Activations