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

    Neuronpedia

    INDEX
    Explanations

    v

    np_max-act · gemini-2.0-flash

    The neuron is triggered by numeric tokens—anything containing digits (literals, register names like v0, constants such as 4, 16, decimal fractions, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     furthermore
    -0.07
    _RANDOM
    -0.06
    وتر
    -0.06
     모르
    -0.06
    ραση
    -0.06
    RadioButton
    -0.06
     college
    -0.06
    уется
    -0.06
    .Hand
    -0.06
    ерта
    -0.06
    POSITIVE LOGITS
     assure
    0.08
    _required
    0.07
     ecosystems
    0.07
     consisted
    0.06
    蘇
    0.06
     architecture
    0.06
     Clair
    0.06
     fixture
    0.06
     систему
    0.06
    XXXX
    0.06
    Act Density 0.004%

    No Known Activations