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

    Neuronpedia

    INDEX
    Explanations

    proper

    np_max-act · gemini-2.0-flash

    phrases related to mathematical structures and their properties.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The main thing this neuron does is pick out technical algebraic‐geometry terms describing properties of maps, sheaves, or spaces (e.g. adjectives like “proper,” “finite,” “flat,” “coherent,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    airobi
    -0.07
     verte
    -0.07
    .root
    -0.06
    _goals
    -0.06
    .percent
    -0.06
     dolphin
    -0.06
     Đài
    -0.06
     جنوبی
    -0.06
     salad
    -0.06
    nad
    -0.06
    POSITIVE LOGITS
    kich
    0.07
    ;z
    0.06
     apt
    0.06
    添
    0.06
     ventured
    0.06
     '[
    0.06
     Specs
    0.06
    生成
    0.06
    dbl
    0.06
     Mushroom
    0.06
    Act Density 0.002%

    No Known Activations