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

    Neuronpedia

    INDEX
    Explanations

    List of numbers

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens that are floating‐point numbers (i.e. numeric strings containing a decimal point and digits after it).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Spain
    -0.07
     intervene
    -0.06
    ían
    -0.06
     imagination
    -0.06
    stru
    -0.06
    miss
    -0.05
     sağ
    -0.05
     Dud
    -0.05
     ajust
    -0.05
     Swansea
    -0.05
    POSITIVE LOGITS
    fillType
    0.07
     answering
    0.07
    환경
    0.07
     Razor
    0.07
    _web
    0.07
     Keeper
    0.07
     volunteer
    0.07
     met
    0.07
    >V
    0.07
     Reception
    0.06
    Act Density 0.037%

    No Known Activations