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

    Neuronpedia

    INDEX
    Explanations

    Greek

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of “Greek” or “Greece,” i.e. references to that country or its demonym.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Jub
    -0.07
    Transport
    -0.07
     malé
    -0.07
    ashboard
    -0.07
     discharge
    -0.07
     NB
    -0.06
    fad
    -0.06
     APR
    -0.06
    ames
    -0.06
     harb
    -0.06
    POSITIVE LOGITS
    Greek
    0.15
     Greek
    0.15
     Greece
    0.14
     Athens
    0.11
     Greeks
    0.10
     gods
    0.08
     ελλην
    0.08
     Brighton
    0.07
    reek
    0.07
     Chinese
    0.07
    Act Density 0.006%

    No Known Activations