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

    Neuronpedia

    INDEX
    Explanations

    -,

    np_max-act · gemini-2.0-flash

    This neuron detects chemical compound names prefixed by Greek letters (e.g., “alpha‐,” “beta‐,” “gamma‐”)

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Diamond
    -0.07
    dob
    -0.07
    iz
    -0.07
    (dist
    -0.06
    Since
    -0.06
    eland
    -0.06
    тов
    -0.06
     INTERFACE
    -0.06
     twist
    -0.06
     While
    -0.06
    POSITIVE LOGITS
     أغسطس
    0.07
     cr
    0.07
     akıl
    0.07
     Quentin
    0.06
     чем
    0.06
    커스
    0.06
     MV
    0.06
    σσότε
    0.06
     Cary
    0.06
     kita
    0.06
    Act Density 0.189%

    No Known Activations