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

    Neuronpedia

    INDEX
    Explanations

    Groups or sets

    np_max-act · gemini-2.0-flash

    The neuron activates on plural nouns (words referring to multiple items, typically ending in “s”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    icates
    -0.07
     Bangalore
    -0.07
    Jay
    -0.07
     Naturally
    -0.06
     recognizable
    -0.06
     StringUtils
    -0.06
    -object
    -0.06
     Transformers
    -0.06
    .Red
    -0.06
    different
    -0.06
    POSITIVE LOGITS
    RYPT
    0.07
    _aut
    0.07
     refl
    0.06
    ีน
    0.06
     Hij
    0.06
     عرض
    0.06
    ΟΣ
    0.06
    urn
    0.06
     좀
    0.06
     viewType
    0.06
    Act Density 0.054%

    No Known Activations